THE UNIVERSITY OF ILLINOIS LIBRARY NATURAL HISTORY SURVEY 5705 ILL V. 5 cop4 ILLINOIS BIOLOGICAL MONOGRAPHS PUBLISHED QUARTERLY UNDER THE AUSPICES OF THE GRADUATE SCHOOL BY THE UNIVERSITY OF ILLINOIS VOLUME III Urbana, Illinois 1916-1917 Editorial Committee Stephen Alfred Forbes ■ William Trelease Henry Baldwin Ward TABLE OF CONTENTS Volume III numi;krs pages 1 Studies on the Factors Controlling the Rate of Regeneration. By Charles Zeleiiy 1-170 (Distributed Novemlier 29, 1916) The Head-Capsule and Mouth-Parts of Diptera. By Alvah Peterson. With 25 plates 171-284 (Distributed December 30, 1016) Studies on Xorth .-Xnierican Polystomidae, .Aspidogastridae, and Parau'.iihistoniidae. r>y Horace Wesley Stunkard. With II plates 285-394 (Distributed May 5. 1917) 4 Color and Color-Pattern Mechanism of Tiger Beetles. By Victor E. Slielford. With 29 black and 3 colored plates 395-532 (Distributed June 30, 191/) ILLINOIS BIOLOGICAL MONOGRAPHS Vol.111 August, 1916 No, I Editorial Committee Stephen Alfred Forbes William Trelease Henry Baldwin Ward Published under the Auspices of the Graduate School by THE University of Illinois Copyright, igi5 By the University of Illinois Distributed November 29, igi6 STUDIES ON THE FACTORS CONTROLLING THE RATE OF REGENERATION CHARLES ZELENY Contributions from the Zoological Laboratory of the University of Illinois, No. 73 732172 TABLE OF CONTENTS PAGE Introduction 7 The Rate of Regeneration from New Tissue Compared with That from Old Tissue 9 The Effect of Successive Removal upon the Rate and Completeness of Regen- eration 26 The Effect of Level of the Cut upon the Rate and Completeness of Regeneration 61 The Change in Rate of Regeneration during the Regenerative Process 108 The Effect of Degree of Injury upon the Rate of Regeneration 136 The Completeness of Regeneration 158 Bibliography 167 RATE OF REGENERATION— ZELEXY INTRODUCTION. The present studies of the factors controlling rate of regeneration are a continuation of previous work on the subject. An advance in knowledge concerning certain of the factors has made possible an exten- sion of the experimental analysis of others. The present studies are therefore closely related. In fact in several cases a single series of iudi- v-iduals has been of value in connection with more than a single factor. The definite determinations of the effect of level of the cut and of the change in rate during the regeneration cycle have been of particular value. The precautions taken to meet the demands of the experiments are not discussed in detail because they have already been given in previous papers. The frog tadpoles (when they can be used) are in all respects more suitable than salamander larvae. When collected late in the fall they can be kept at a fairly constant size and the results obtained under these conditions are not complicated with growth phenomena. They have proved to be remarkably uniform in several series. The salamander larvae on the other hand vary in rate of regeneration from day to day. The factors involved in this fluctuation were not discovered and could not be remedied but may be related in some way to the fact that these animals require living active food and the feeding reactions are therefore more complicated tlian in frog tadpoles and more subject to disturbance. In regard to certain factors, such as the degree of injury, in which expected differences in rate are slight the writer has felt that he might be biased in making the measurements and in a number of cases this work was therefore delegated to a person who had no preconceptions concerning the result. In making averages elimination of individual cases is avoided except for a few very aberrant values. Such exceptional values are in every case however included in the tables. In many cases where only slight differences are to be expected several different kinds of comparisons are made .so as to bring out the correct relation as completely as possible. As in the past all data obtained by the writer on the particular factors in question are given. The practice of selective elimination would be dangerous because of the large value of factors not at present under experimental control. 8 ILLINOIS BIOLOGICAL MONOGRAPHS [8 DiscussioDS of the results of other workers are included in the pre- vious papers and need not be repeated here. The principal need at present seems to be an extension of knowledge of these factors by multiplying the number of series of carefully controlled experiments. "While it would be interesting to know why a particular series differs from others with respect to a certain factor it is not always possible to discuss the matter profitably in the absence of evidence as to all the factors concerned. Particular emphasis must be laid on the fact that in connection with some at least of the factors it has been possible to make out very definite quantitative relations. These have been checked up in a number of cases by agreement between separate series of experiments. The success in this direction has made it very probable that with a more accurate control of external conditions there will be a considerable further advance in our knowledge of the factors controlling rate of regeneration. RATE OF REGEXERATION—ZELESY PART I THE RATE OF REGENERATION FROM NEW TISSUE COM- PARED WITH THAT FROM OLD TISSUE In comparing first and second regenerations from the same level one of the diflSculties that presents itself is tlie impossibility of making the second cut exactly in the path of the first. This is true not only because of the error in manipulation but also because the old and the new tissues become intermingled and do not retain a distiuct dividing line. At the cut surface there is old tissue alone, old and new tissue, or new tissue alone according as the second cut comes inside of the first level, exactly at the level, or outside of it. The experiments about to be described were devised with a view to the testing of the relative rates from old and from new tissue. Other factors being eliminated, are new cells which are recently produced in a regenerating part able to carry on a repetition of the process more expeditiously than old cells which have not been directl}^ concerned in such a process? There has been no selective elimination of data. As in former papers of a similar character all the data obtained by the author on the topic at hand are included. Experiment I Series 3628-3675 Tadpoles of Rana clamitans with an average length of 33.4 mm. were used. They were fed just enough to keep them in good condition without much growth. AH were collected at one time in a single pool and during the course of the experi^ient factors apart from the one under investigation were made as nearly alike as possible. This elimination of outside factors was facilitated by subdividing the tad- poles into sets of two each, the two indiNnduals of a set being exactly alike except for the factor under consideration and one being used for regeneration from old tissue and the other for regeneration from new tissue. 10 ILLINOIS BIOLOGICAL MONOGRAPHS [10 Within each set the tail of tadpole 1 was removed at B (Fig. 1) and the tail of tadpole 2 at A. The distance between A and B was 2 or 3 mm. After 21 days of regeneration the second operation on both tadpoles came between A and B and tlierefore in old tissue in tadpole 1 and in new tissue in tadpole 2. This procedure, insuring ^ B Figure 1. Outline of tadpole of Rana claiiiitans- Individuals used for re- generation from old tissue have the original removal level at B and the second level at A. Individuals used for regeneration from nevif tissue have the original removal level at A and the second level at B. Regenerations from the second levels are compared. approximately the same level in the two cases, is necessary because level of the cut has a great influence upon rate of regeneration. Eleven pairs of individuals were used in the comparison. The precautions taken to eliminate possible error are treated fully elsewhere for similar cases (Zeleny 1909a, 1909b). The data are given in Table 1. The removed tail lengths are the lengths of the original removed portions of the tail plus or minus the EXPLANATION OP TABLE I. Note 1. The removed length is the length of the original removed portion of the tail plus or minus the distance of the new cut surface from the dividing line between the old and new tissue. Note 2. The lengths as given are the living lengths. Measurements were made on material killed in Gilson's mercuro-nitric mixture and preserved in 85% alcohol. Sets I and IX were measured both when alive and after killing and preserving. From tkem the shrinkage coefBcient was obtained and this made possible the reduction of all the data to the living basis. Note 3. The specific amount regenerated in any case is the amount regen- erated per unit of removed length. Note 4. The average includes only the sets in which both individuals are present. 11] RATE OF REGEXERATION—ZELEXy TABLE I. Series 3628-3675 Old or Nev tissue at out surface Old new old II new old III new old IV new old V new old VI new old VII new old VIII new old IX new old X new old XI new Averag e of old Averag e of new Old— a dead New — i ihea i Cata- log Dumber Total length mm. moved length mm. Old — Times ahead New — Times ahead 3628 3629 3633 3632 3636 3637 3641 3640 3645 3644 3649 3648 3652 3653 3656 3657 3660 3661 3668 3669 3672 3673 38.0 24.1 39.2 24.6 35.7 23.2 33.8 22.1 35.8 23.1 38.4 32.9 25.0 20.8 31.4 20.4 37.5 23.8 42.8 29.2 37.0 25.6 35.9 23.3 31.3 20.8 29.0 19.2 31.8 21.1 33.0 22.0 26.5 17.0 29.4 19.0 31.1 20.8 32.4 21.8 24.4 15.8 28.5 18.1 32.9 21.5 34.0 22.2 13.2 12.8 12.3 10.2 12.8 11.9 11.3 9.3 11.5 15.1 11.2 9.9 13.2 11.7 9.6 8.7 9.2 8.9 8.6 8.7 11.3 10.7 Regen- I Specific erated length length regen- mm . I erated 2.2 2.3 Regen- I Specific erated I length length I regen- mm I erated 2.0 1.8 2.0 2.4 1.7 2.2 2.2 2.3 2.3 2.1 2.4 2.7 2.5 2.0 1.6 1.8 2.7 1.9 2.16 2.15 0.17 0.18 O.IG 0.18 0.16 0.20 0.15 0.24 0.19 0.15 0.21 0.21 3.5 3.1 3.25 3.5 3.1 3.1 0.27 0.24 0.25 0.29 0.28 0.31 0.26 0.18 0.23 0.26 0.23 0.17 0.20 0.31 0.22 0.196 0.204 31/2 6/2 3.6 3.5 3.0 2.3 2.5 3.5 3.5 0.07 0.36 0.34 0.25 0.28 0.41 0.40 0.303 0.310 12 ILLINOIS BIOLOGICAL MONOGRAPHS [12 distances of the new cut surface from the dividing line between the old and the new tissue. The regenerated lengths as given are the living lengths. Measm-ements were made on material killed in Gilson's mer- curo-nitrie fluid and preserved in 85% alcohol. Sets I and IX were measured both wlien alive and after killing and preserving. From them the shrinkage coefficient was obtained and this made possible the reduc- tion of all tlie data to the living basis. The averages include only the sets in which both individuals are present. The specific amount of regeneration is the amount regenerated per unit of removed length. It has been shown that within wide limits this is a constant if the only- variable in the experiment is the amount removed. This statement holds for all levels in the present experiment. The table shows that the average amount regenerated at the end of six days is 2.16 mm. from the old tissue levels and 2.15 mm. from the new tissue levels. The new tissue levels however rep- resent the shorter amount removed, 10.7 mm. as opposed to 11.3 for the old tissue levels. This gives an average specific rate of 0.204 for the new levels and 0.196 for the old levels. The difference is proba- bly not significant. The individual specific amoimts in pairs, putting the old tissue first and the new tissue second in each case, are 0.17 and 0.18, 0.16 and 0.18. 0.16 and 0.20, 0.15 and 0.24, 0.19 and 0.15, 0.21 and 0.21, 0.18 and 0.23, 0.26 and 0.23, 0.17 and 0.20, and 0.31 and 0.22. The old tissue is ahead three times, the new six times and there is a tie in one case. At the end of eight days the result is similar. There is a slight advantage in favor of the new tissue level but this cannot be considered as significant. The average amount regenerated is 3.19 mm. from old tissue levels and 3.12 mm. from new tissue levels. The specifile amount regenerated is 0.303 for the old and 0.310 for the new level. The individual amounts by pairs putting the old tissue level first as before are 0.27 and 0.24, 0.25 and 0.29, 0.28 and 0.31, 0.36 and 0.34, 0.25 and 0.28, and 0.41 and 0.40. Each level is ahead of the other in three of the six cases. Experiment II Series 3676-3765 Tadpoles of Rana clamitans with an average length of forty mm. were used. The experiment was designed for a study of the effect of successive removal on the rate of regeneration but incidentally furnishes valuable data for the present problem. In removing the re- generated portion, the cut in most cases did not come exactly at the border. In some cases it was too near the base of the tail and therefore the cells at the cut surface were old unregenerated cells. In other cases 13] RATE OF REGEXERATIOX—ZELENY 13 it was too near the tip of the tail and the cells at the cut surface were newly regenerated ones. The operations were at different levels in dift'erent individuals but the determination of the specific amounts of regeneration accoi'ding to the method given in the explanation of Experiment I eliminates these differences within wide limits. It does not hold when the level of the cut is very near the tip or near the base of the tail. In the present experiment the specific amount is a fair constant for all removed lengths of over 4 mm. The individuals with a removed length of less than 4 mm. are therefore treated separately. Likewise it does not hold for the first few days of regeneration during which regenera- tion is confined to active migration of cells over the cut surface without any new formation by cell division. Separate comparisons are made at 4, 6, 8, 10, 121/4, 18 and 56 days of regeneration. The data are given in Tables 2 to 17. Taking first the cases with a removed length of over 4 mm. there is at four days a specific amount of 0.043 for old tissue and of 0.045 for new tissue. At six days the amounts are respectively 0.135 and 0.143, at eight days 0.216 aiid 0.224, at ten days 0.292 and 0.293, at twelve and a half days 0.331 and 0.337, at eighteen days 0.352 and 0.348, and at fifty-six days 0.345 and 0.346. The two are approximately equal though in six out of the seven cases the new tissue is ahead. The average difference in favor of the new tissue is 0.003. For removed amounts of less than 4 mm. the data are un- satisfactory because there are only three individuals with regeneration from new tissues. The data are however of value in comparison with the others. The specific amounts at the different days, again putting the old tissue first in each case, are 0.119 and 0.160 for four days, 0.317 and 0.327 for six days, 0.444 and 0.467 for eight days, 0.506 and 0.520 for ten daj'S, 0.517 and 0.517 for twelve and a half days, 0.501 and 0.507 for eighteen days, and 0.475 and 0.325 for fifty-six days. In the last the absorption of the tail had begun before the measurement was made and the comparison is therefore not valid for our purposes. In the first, 0.119 for old and 0.160 for new at four daj'S, the great difference between individual cases on eacli side makes a comparison of doubtful validity. There are other data however which make it probable that tlio initial migration of the cells takes place more rapidly from new tlum from old tissue. For the other levels there is on the average a slight difference (0.011) in favor of the regeneration from new tissue. With but a single exception, which is a tie, the new tissue is ahead of the old. The differences favoring the new tissue are greater than those for the larger removals. This again may be due to the fact that a larger percent- ILLINOIS BIOLOGICAL MONOGRAPHS [14 age of the regenerated material is derived from the old by migration and a smaller percentage by cell division. The data unfortunately are based on such a small number of individuals, especially in the case of new tissue levels, that too much stress should not be laid on the differences. TABLE 2 Series 3676-3765 Over 4 millimeters removed Regeneration; 4 days Note 1. No. 3734 is left out in making up the averages because its specific amount from six days of regeneration on is very much in excess of that of any of the others. A probable explanation is that the end of the tail in this indi- vidual had been removed and regeneration had just started when the present operations were begun. If this is true it belongs to a longer removed length than indicated and the specific rate is wrong. Besides a highly exceptional Individual even if not explained should be left out in determining the average value. 15] RATE OF REGEXERATION—ZELENY Series 3676-3765 TABLE 3 Over 4 millimeters removed Regeneration: 6 days Series 3676-3765 TABLE 4 Over 4 millimeters removed Oldt ssue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated mm. Specific length regen- erated 3720 4.7 0.84 0.18 3756 4.8 1.0 0.21 3648 5.5 0.7 0.13 3751 6.7 1.1 0.16 3715 7.9 0.9 0.11 3697 7.3 1.2 0.16 3757 8.0 1.2 0.15 3721 8.5 1.3 0.15 3694 8.7 1.5 0.17 3733 8.5 1.0 0.12 3685 9.3 1.2 0.13 3734 8.5 2.1 0.25 3686 14.5 2.1 0.14 3739 9.4 1.0 0.11 3753 16.8 2.0 0.12 3722 12.5 1.6 0.13 3723 18.4 2.3 0.12 3716 12.7 1.7 0.13 3699 21.0 2.2 0.10 3698 12.9 1.5 0.12 3759 15.5 2.0 0.13 3705 17.6 2.0 0.11 3717 17.6 2.6 0.15 3687 19.7 2.6 0.18 Average 0.135 Average 1 0.143 Regeneration: 8 days Old tissue New tissue Catalog number Length removed mm. Length regen- erated mm. Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3720 4.7 1.1 0.23 3756 4.8 1.3 0.27 3684 . 5.5 1.2 0.22 3751 6.7 1.7 0.25 3715 7.9 1.7 0.22 3697 7.3 1.7 0.23 3757 8.0 2.1 0.26 3721 8.5 1.9 0.22 3694 8.7 2.2 0.25 3733 8.5 1.9 0.22 3685 9.3 1.9 0.20 3734 8.5 3.1 0.36 3686 14.5 3.4 0.23 3739 9.4 1.8 0.19 3753 16.8 2.5 0.15 3722 12.5 2.6 0.21 3723 18.4 3.7 0.20 3716 12.7 2.4 0.19 3699 21.0 4.3 0.20 3698 12.9 3.3 0.26 3759 15.5 3.0 0.19 3705 17.6 3.6 0.20 3717 17.6 3.6 0.20 3687 19.7 5.6 0.28 Average 0.216 Average 0.224 ILLLXOIS BIOLOGICAL MONOGRAPHS TABLE 5 Series 3676-3765 Over 4 millimeters removed Regeneration: 10 days Old tissue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3720 4.7 1.3 0.28 3756 4.8 1.7 0.35 3684 5.5 1.4 0.25 3751 6.7 2.1 0.31 3715 7.9 2.3 0.29 3697 7.3 2.2 0.30 3757 8.0 2.8 0.35 3721 8.5 2.3 0.27 3694 8.7 3.2 0.37 3733 8.5 2.4 0.28 3685 9.3 2.3 0.25 3734 8.5 4.5 0.53 3686 14.5 4.8 0.33 3739 9.4 2.4 0.26 3753 16.8 3.8 0.23 3722 12.5 3.6 0.29 3723 18.4 5.3 0.29 3716 12.7 3.4 0.27 3699 21.0 5.9 0.28 3698 12.9 4.3 0.33 3759 15.5 4.2 0.27 3705 17.6 4.8 0.28 3717 17.6 5.2 0.30 3687 19.7 6.0 0.30 Average 0.292 Average 0.293 Series 3676-3765 TABLE 6 Over 4 millimeters removed Regeneration; 12-13 days Old tissue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3720 4.7 1.3 0.28 3756 4.8 1.8 0.37 3684 5.5 1.4 0.25 3751 6.7 2.4 0.36 3715 7.9 2.6 0.33 3697 7.3 2.4 0.33 3757 8.0 3.1 0.39 3721 8.5 2.6 0.31 3694 8.7 3.4 0.39 3733 8.5 2.6 0.31 3685 9.3 2.8 0.30 3734 8.5 5.7 0.67 3G86 14.5 5.3 0.37 3739 9.4 3.0 0.32 3753 16.8 5.2 0.31 3722 12.5 3.9 0.31 3723 18.4 6.5 0.35 3716 12.7 4.2 0.33 3699 21.0 7.1 0.34 3698 12.9 5.0 0.39 3759 15.5 4.8 0.31 3705 17.6 6.4 0.36 3717 17.6 6.0 0.34 3687 19.7 6.6 0.34 Average 0.331 Average 0.337 RATE OF REGENERATION — ZELENY Series 3676-3765 TABLE 7 Over 4 millimeters removed Regeneration: 17-18-19 days Old tissue New tissue Catalog number Length removed mm. Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated mm. Specific length regen- erated 3720 4.7 1.3 0.28 3756 4.8 1.6 0.33 3684 5.5 1.5 0.27 3751 6.7 2.5 0.37 3715 7.9 2.6 0.33 3697 7.3 2.3 0.32 3757 8.0 3.2 0.40 3721 8.5 2.3 0.27 3694 8.7 3.4 0.39 3733 8.5 2.6 0.31 3685 9.3 3.0 0.32 3734 8.5 6.4 0.75 3686 14.5 5.2 0.36 3739 9.4 2.9 0.31 3753 16.8 6.4 0.38 3722 12.5 3.5 0.28 3723 18.4 8.1 0.43 3716 12.7 5.1 0.40 3699 21.0 7.5 0.36 3698 12.9 5.4 0.42 3759 15.5 6.7 0.43 3705 17.6 6.2 0.35 3717 17.6 6.7 0.38 3687 19.7 7.0 0.36 Average 0.352 Average 0.348 TABLE 8 Series 3676-3765 Over 4 millimeters removed Regeneration: 55-56-57 days Old tissue New tissue Catalog number Length removed mm. Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3720 4.7 1.3 0.28 3756 4.8 — 3684 5.5 1.4 0.25 3751 6.7 2.3 0.34 3715 7.9 2.8 0.35 3697 7.3 2.1 0.29 3757 8.0 3.1 0.39 3721 8.5 2.2 0.26 3694 8.7 — 3733 8.5 2.8 0.33 3685 9.3 2.6 0.28 3734 8.5 6.6 0.78 3686 14.5 — 3739 9.4 — 3753 16.8 7.1 0.42 3722 12.5 4.2 0.34 3723 18.4 8.3 0.45 3716 12.7 4.4 0.35 3699 21.0 7.2 0.34 3698 12.9 5.4 0.42 3759 15.5 6.6 0.43 3705 17.6 6.0 0.34 3717 17.6 6.4 0.36 3687 19.7 — Average 0.345 Average 0.346 ILLINOIS BIOLOGICAL MONOGRAPHS [18 Series 3676-3765 TABLE 9 Over 4 millimeters removed Summary Tables 2 to 8 Table number Days of regeneration Old tissue Specific length of regeneration New tissue Specific lengtli of regeneration Old ahead New ahead 2 4 0.043 0.045 0.002 3 6 0.135 0.143 0.008 4 8 0.216 0.224 0.008 5 10 0.292 0.293 0.001 6 12,13 0.331 0.337 0.006 7 17,18,19 0.352 0.348 0.004 8 55, 56, 57 0.345 0.346 0.001 Average 0.003 TABLE 10 Series 3676-3765 Less than 4 millimeters removed Regeneration: 4 days Old tissue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3676 1.3 0.27 0.27 3696 2.1 0-48 0.23 36S2 1.6 0.18 0.11 3749 2.8 0.30 0.11 3730 1.6 0.39 0.24 3750 3.5 0.48 0.14 3754 1.6 0.06 0.04 3718 2.1 0.06 0.03 3731 2.7 0.15 0.06 3713 2.8 0.36 0.13 3719 3.1 0.36 0.12 3701 3.2 0.42 0.13 Average 0.119 Average 0.160 19] RATE OF REGE.XERATIOX—ZELEXV TABLE 11 Series 3676-3765 Less than 4 millimeters removed Regeneration: 6 days Old tissue New tissue Catalog number Length removed mm. Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3676 1.3 0.6 0.46 3696 2.1 0.85 0.40 3682 1.6 0.6 0.37 3749 2.8 0.6 0.21 3730 1.6 0.75 0.47 3750 3.5 1.3 0.37 3754 1.6 0.55 0.34 3718 2.1 0.45 0.21 3731 2.7 0.5 0.19 3713 2.8 0.8 0.29 3719 3.1 0.84 0.27 3701 3.2 0.8 0.25 Average 0.317 Average 0.327 Series 3676-3765 TABLE 12 Less than 4 millimeters removed Regeneration: 8 days Old tissue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated 0.69 Catalog number Length removed Length regen- erated mm. Specific length regen- erated 3676 1.3 0.9 3696 2.1 1.0 0.48 3682 1.6 0.9 0.56 3749 2.8 1.2 0.43 3730 1.6 0.9 0.56 3750 3.5 L7 0.49 3754 1.6 0.9 0.56 3718 2.1 0.7 0.33 3731 2.7 0.8 0.29 3713 2.8 0.9 0.32 3719 3.1 1.1 0.35 3701 3.2 1.1 0.34 0.444 Average Average 0.467 20 ILLINOIS BIOLOGICAL MONOGRAPHS [20 Series 3676-3765 TABLE 13 Less than 4 millimeters removed Regeneration: 10 days Old tissue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3676 1.3 0.9 0.69 3696 2.1 1.1 0.52 3682 1.6 1.0 0.62 3749 2.8 1.4 0.50 3730 1.6 0.9 0.56 3750 3.5 1.9 0.54 3754 1.6 1.1 0.69 3718 2.1 1.0 0.48 3731 2.7 1.0 0.37 3713 2.8 0.9 0.32 3719 3.1 1.4 0.45 3701 3.2 1.2 0.37 Average 0.506 Average 0.520 Series 3676-3765 TABLE 14 Less than 4 millimeters removed Regeneration: 12-13 days Old tissue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3676 1.3 0.9 0.69 3696 2.1 1.0 0.48 3682 1.6 1.0 0.62 3749 2.8 ; 1.4 0.50 3730 1.6 0.9 0.56 3750 3.5 2.0 0.57 3754 1.6 1.2 0.75 3718 2.1 1.0 0.48 3731 2.7 1.0 0.37 3713 2.8 0.9 0.32 3719 3.1 1.4 0.45 3701 3.2 1.3 0.41 Average 0.517 Average 0.517 RATE OF RECEKERATIOX — ZELESY TABLE 15 Series 3676-3765 Less than 4 millimeters removed Regeneration: 17-18-19 days Old tissue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated Catalog number Length removed Length regen- erated Specific length regen- erated 3676 1.3 0.9 0.69 3696 2.1 1.0 0.48 3682 1.6 1.0 0.62 3749 2.8 1.3 0.47 3730 1.6 0.9 0.56 3750 3.5 2.0 0.57 3754 1.6 1.2 0.75 3718 2.1 0.5 0.24 3731 2.7 — 3713 2.8 0.9 0.32 3719 3.1 1.3 0.42 3701 3.2 1.3 0.41 Average 0.501 Average 0.507 TABLE 16 Series 3676-3765 Less than 4 millimeters removed Regeneration: 55-56-57 days Oldti ssue New tissue Catalog number Length removed Length regen- erated Specific length regen- erated 0.54 Catalog number Length removed Length regen- erated Specific length regen- erated 3676 1.3 0.7 3696 2.1 0.7 0.33 3682 1.6 1.1 0.69 3749 2.8 0.9 0.32 3730 1.6 0.7 0.44 3750 3.5 — 3754 1.6 1.1 0.69 3718 2.1 — 3731 2.7 — 3713 2.8 0.5 0.18 3719 3.1 — 3701 3.2 1.0 0.31 Average 0.475 Average 0.325 22 ILLIXOIS BIOLOGICAL MONOGRAPhlS TABLE 17 Series 3676-3765 Less than 4 millimeters removed Summary Tables 10 to 16 Table number Days of regeneration Old tissue Specific length of regeneration New tissue Specific length of regeneration Old ahead New ahead 10 4 0.119 0.160 0.041 11 6 0.317 0.327 0.010 12 8 0.444 0.467 0.023 13 10 0.506 0.520 0.014 14 12,13 0.517 0.517 0.000 0.000 15 17,18, 19 0.501 0.507 0.006 16 55, 56, 57 0.475 0.325 0.150 Average ■ 0.011 Note 1. Because of the great variability in the data the average for the four-day period is not of much value and is therefore not included In the grand average. Note 2. The absorption of the regenerated portion of the tail was pro- ceeding so rapidly by the fifty-fifth day of regeneration that this average should not be included in the grand average. Experiment III Series 3557-3624 This experiment was planned for a study of the effect of repeated removal and regeneration upon the rate of metamorphosis but it yields data of value for the present problem. Tadpoles of Rana clamiians with an average length of about 40 mm. were used. In some cases the cuts were made inside of the first level and therefore in old tissue and in other cases outside of tlie first level and therefore in new tissue. The data include third, fourth and fifth successive regenerations. The time of regeneration is 37 days for the third and 36 days for the fourth and for the fifth regenerations. The length of time is more than sufficient for the completion of the process of regeneration in so far as it is completed. The data therefore do not serve for the rate but for the completeness of regeneration from old as compared with new levels. Approximately one-half of the original tail length was removed but measurements were not made of individual removed lengths, so that 23] RATE OF REGEXERATIOX—ZELENY 23 specific rates of regeneratiou can not be calculated. However the re- moved lengths were so nearly alike as to make the regenerated lengths of value in direct comparison. The data are given in Table IS. The average length of the third regeneration is 7.9 mm. for both the old and the new tissue basis. For the fourth regeneration the value from old tissue is 5.3 mm. and from new tissue 5.5 mm. The corresponding values for the fifth regeneration are 6.6 mm. and 5.9 mm. Averaging the individual eases for all three regenerations the old tissue average is 6.5 mm. and the new tissue average 6.6 mm., an advantage in favor of the latter of 0.1 mm. This difference can not be consid- ered as significant, especiall.y since for the individual regenerations the two levels give equal regenerated lengths for the third, the new is slightly ahead at the fourth and the old is ahead at the fifth. On the whole the data for Experiment III agree witli those for Experiments I and II. There is no striking difference between com- pleteness of regeneration from old and from new tissue levels, though a small difference favoring the latter persists in practically all the comparisons. TABLE 18 Rana clamitans Series 3557-3624 Regenerated tail length from new tissue compared with that from old tissue during the third, fourth and fifth regenerations Third regeneration 37 Days Fourth regeneration Fifth regeneration Third, fourth and fifth regenerations combined 36 Days 36 Days Old New Old New Old New- Old New tissue tissue tissue tissue tissue tissue tissue tissue 2.0 5.7 4.4 4.9 4.7 5.2 6.8 6.6 4.5 .5.4 5.5 5.7 6.9 8.0 4.8 5.5 6.2 5.9 7.5 8.3 4.9 6.1 6.5 6.8 7.9 9.3 5.0 7.1 9.0 9.7 ."..1 7.2 9.4 5.S t;.i 7.3 7.9 8.0 Average in mm. 7.9 7.9 5.3 5.5 6.6 5.9 6.5 6.6 Difference in mm. 0.0 0.0 +0.2 4-0.7 -fO.1 24 ILLINOIS BIOLOGICAL MONOGRAPHS [24 Discussion While tlie knowledge of the relative rates of regeneration for old and new tissue is essential for accurate determination of other factors its main interest is in its bearing on the question of the character of control of the process of regeneration. Evidence from a great many directions points toward the conclusion that regeneration is not wholly a direct response of the injured cells at the cut surface nor of those in the immediate neighborhood of the cut surface. It is more and more evident that conditions in parts of the body remote from the injured region are involved. If rate of regeneration were determined wholly by the character of the cells at the cut surface we would expect that cells in process of active proliferation, such as those that are starting to build up a new tail, would respond much more promptly than those which have become more highly diffei-entiated and hence more stable. Regenerating cells ought to furnish a much better basis than old ones. "We find however that there is no striking difference in the two cases. Regeneration proceeds at approximately the same rate whether old or new cells have furnished the basis for the new material. It is true that the data show on the average a slight advantage in favor of the new tissue, especially during the early periods, but this advantage is small and it is doubtful Avhether it can be considered as significant. There is some evidence that the earliest stages of regeneration, those due to cell migration exclusively, are more rapid from new than from old tissue. If this evidence is reliable an explanation is found for the slight advan- tage in favor of the new tissue at later periods. Summary 1. A comparison of the rate of regeneration in tadpoles of Raiia clamitans in eases where there are newly regenerated cells at the cut surface with those in which only old cells are present shows, on the whole, little difference between the two. 2. The slight difference favors the new cells but may not be significant. 3. In Experiment I the specific length of regeneration at the end of 6 days was 0.196 from old tissue and 0.204 from new tissue. 4. In the same experiment at the end of 8 days the specific length from the old was 0.303 and from the new 0.310. 5. In Experiment II the general result was similar to that in Experiment I. The amounts of regeneration in the two cases are very nearly equal and the slight difference is in favor of the new tissue. 6. Experiment III shows that as regards completeness of regen- 25] RATE OF REGEXERATION—ZELENY 25 eration there is again essential similarity between the old tissue and the new tissue levels. 7. The result strengthens the view that the rate of regeneration is controlled in large part by factors not inherent in the character or con- dition of the cells near the cut surface. 8. In the case of the earliest stages, those in which there is cell migration but no cell division, there is some evidence that the rate of regeneration may be greater from new than from old tissue. 26 ILLIXOIS BIOLOGICAL MOXOGRAPHS [26 PART II THE EFFECT OF SUCCESSIVE REMOVAL UPON THE RATE AND COMPLETENESS OF REGENERATION One of the most ioteresting facts in connection with regeneration is the ability to replace a part after repeated removal. The present set of experiments was made in continuation of jarevious studies of the effect of successive removal upon the rate of regeneration (Zeleny 1907, 1908, 1909). The earlier studies show that as a rule the rate of regen- eration following a first removal is no greater than that following second and latei^reuiovals if the effect of age is eliminated. Where a difference exists it seems to be in favor of the later regenerations. The matter is of very great interest in connection with general problems of development and particularly in connection with 'the ques- tion as to the existence or non-existence of a necessary limit to the amount of living substance that a single individual may produce during its life cycle. Does the production of a group of tissues use up a part of a certain store of developmental energy or of developmental factors possessed by the individual or is this store inexhaiistible or perchance even increased by exercise of the functioia? These questions warrant more extended study especially in view of the additional analysis that has been made of other factors controlling the rate of regeneration. The paper includes all the unpublished data that have been obtained on the problem at hand. In general these data support the conclusions previously reached. The descriptions of the individual experiments will first be given and they will be followed bj- a discussion of the general results. Experiment I Rana clamitans Skries 3628-3675 Material and Method The tadpoles were collected on December 9, 1911. At the time of the operation on December 20 the average total length Avas 33.0 mm. and the average tail length 21.6 mm. Forty-eight individuals were divided into twelve sets of four each. The four indi- viduals of a set are called a, h, c, and d. Approximately one-half in length of the tail was removed by a transverse cut in c and d. After 21 days the regenerated portion of the tail was removed. In individual c the second cut came inside of the border liue between old and new 27] RATE OF REGEKERATIOS — ZELENY 27 tissue and iu individual d it came outside of that liue. Of the two indi- viduals available for second regeneration in each set, the one with the cut nearer to the tip of the tail was chosen as individual c and the other as individual d. In this way the second regeneration levels were equal- ized. A first removal of a half of the tail was made in individuals a and h at the same time that the second removal was made in c and d. A direct comparison of the rate of the second regeneration with that of the first was thus made possible without the complication due to internal factors such as difference in age, or external factors such as temperature and food. Measurements of regenerated lengths were made at the end of six and of eight days, other experiments having shown that the period of most rapid growth comes at about this time. Elsewhere there is a comparison of the rate of regeneration from new tissue with that from old tissue. Here the chief concern is the comparison of the rate of the second regenerations, including both old tissue and new tissue levels, with first regenerations. Data The results of the experiment are given in Table 19 for six- day regenerations and in Table 20 for eiglit-day regenerations. At the end of six days the average length of first regenerations is 2.01 mm. and of second regenerations 2.18 mm. Tlie first exceeds the second in two cases, the second exceeds the first in eight and one is tied. The corresponding average specific amounts are 0.194 and 0.205. In five eases the first exceeds the second and in six the second exceeds the first. At eight days the average length of the first regenerations is 3.06 mm. and of the second 3.42 mm. The first exceeds the second iu three sets and the second exceeds the first in seven sets. The corresponding aver- age specific amounts are 0.298 and 0.323. In four the first exceeds the second regeneration and in six the second exceeds the first. Compai'ing the first regenerations on the one hand with second regenerations from old tissue and on the otlier hand with second regen- erations from new tissue it is found, including only complete sets, that at the end of six days the average first regeneration length is 2.01 mm. while that of the second from new tissue is 2.15 mm. and from old tissue 2.16 mm. The corresponding average specific amounts are 0.194 for first regenerations and 0.196 for second regenerations from old tissue and 0.204 for second regenerations from new tissue. At eight days the first regeneration lengths average 3.06 mm. while second regenerations from old tissue average 3.19 and those from new tissue 3.12. The corresponding specific lengths are 0.298 for first regen- erations and 0.303 for second from old tissue and 0.310 for second from new tissue. 28 ILLINOIS BIOLOGICAL MOXOGRAFHS TABLE 19 Rana clamitans Series 3676-3765 Comparison of first and second regenerations Age factor eliminated Six Days [28 Series Regen- eration Total length Tail length Length moved l.engtl- re gen erated Specific length regen erated Aver- age length regen- erated Aver- age speci6c length regen- erated 1 individual a 35.5 23.1 11.5 1.9 0.17 ^ individual b 34.5 21.9 24.1 9.4 1.8 2.2 0.19 1.85 0.180 2 c from old tissue 38.0 13.2 0.17 d from new tissue 39.2 24.6 12.8 2.3 0.18 2.25 0.175 ' 1 a b 34.5 33.9 23.0 22.2 ' ' ' ' II 9.6 1.7 0.18 1.70 O.ISO 2 c old 35.7 23.2 12.3 2.0 0.16 d new 33.8 22.1 10.2 1.8 0.18 1.90 0.170 1 a 36.2 23.3 9.7 1.7 0.18 b 34.1 22.5 10.9 1.9 0.17 1.80 0.175 III 2 c old 35.S 23.1 12.8 2.0 0.16 d new 38.4 25.0 11.9 2.4 0.20 2.20 0.180 1 a 33.1 21.2 12.6 '2.-2- 0.17 b 32.4 20.8 10.2 1.7 0.17 1.95 0.170 IV J 2 c old 32.9 20.8 11.3 1.7 0.15 d new 31.4 20.4 9.3 2.2 0.24 1.95 0.195 1 a 40.8 27.3 11.9 2.1 0.18 b 39.4 26.4 12.7 2.2 0.17 2.15 0.175 V 2 c old 37.5 23.8 11.5 2.2 0.19 d new 42.8 29.2 15.1 2.3 0.15 2.25 0.170 1 a 37.0 24.5 10.2 1.9 0.19 b 35.7 24.6 12.9 2.2 0.17 2.05 0.180 VI 2 c old 37.0 25.6 11.2 2.3 0.21 d new 35.9 23.3 9.9 2.1 0.21 2.20 0.210 1 a 31.2 20.1 11.9 2.1 0.18 b 28.0 18.6 8.4 2.0 0.24 2.05 0.210 VII 2 c old 31.3 20.8 — d new 29.0 19.2 9.2 2.4 0.26 2.40 0.260 29] RATE OF RECEXERATIOX—ZELEXy 29 TABLE 19 (Contiuued) Series 1 Re«en- eration 1 VIII 2 1 IX 2 1 X 2 1 XI 2 1 XII 2 1 Average 2 a b c old d new a b c old d new a b c old d new a b c old d new a b c old d new Total length Tail length Length re- moved Length regen. erated Specific length regen- erated .\ver- age length regen- erated 28.7 32.0 18.5 21.5 21.1 22.0 10.0 8.5 2.1 2.2 2.4 2.7 0.21 0.26 2.17 31.8 33.0 13.2 11.7 0.18 0.23 2.55 29.8 26.9 19.1 17.0 10.1 10.7 2.3 2.3 0.23 0.22 2.30 26.5 29.4 17.0 19.0 9.6 8.7 2.5 2.0 0.26 0.23 2.25 32.1 32.4 21.5 21.5 22.4 19.8 12.0 10.1 2.3 2.0 0.19 0.20 2.15 32.7 30.0 — 30.9 30.1 20,9 20.2 10.2 9.4 2.0 1.8 0.20 0.20 1.90 31.1 32.4 20.8 21.8 9.2 8.9 1.6 1.8 0.17 0.20 1.70 28.0. 26.3 18.0 16.4 11.0 10.4 2.2 2.1 0.20 0.20 2.15 24.4 28.5 15.4 18.1 8.6 8.7 2.7 1.9 0.31 0.22 2.30 32.7 21.4 10.6 2.01 33.4 21.8 10.9 2.18 0.185 0.265 0Tl94 The data as a whole show an advantage in favor of the second regeneration as compared with the first. This is seen not only when the direct regenerated lengths are taken but also when the specific amounts are used. Elsewhere it is shown that the specific amount of regeneration is independent of the level of the cut and therefore a constant within the limits of removal as used in this experiment. The specific amount 30 ILLINOIS BIOLOGICAL MONOGRAPHS [30 determinations are therefore more accurate for our purposes than the direct values of length regenerated. Tlie first regeneration is slight!}' below the second not only in case the latter is from new cells but also in case it is from old cells. The difference between first and second regenerations therefore can not be due entirely to the presence in the former of cells wliicli are already undergoing regeneration. TABLE 20 Rana clamitans Series 3628-3675 First and second regenerations compared Age factor eliminated Eight days Re- gener ation Length removed Length regen- erated Specific length regen- erated .\verage length regen- erated Average specific length regen- erated 1 individual a 11.5 3.1 0.26 individual b 9.4 2.5 0.27 0,28 0.265 2 c from old tissue 13.2 3.5 0.27 d from new tissue 12.8 3.1 0.24 3.30 0.255 1 a — ' b 9.6 2.1 0,22 2.10 0.220 2 c old 12.3 3.4 0.28 d new 10.2 — 3.40 0.280 1 a 9.7 2.7 0.28 b 10.9 3.4 0.31 3.05 0.295 2 c old 12.8 3.25 0.25 d new 11.9 3.5 0.29 3.37 0.270 1 a 12.6 3.25 0.26 2 b 10.2 3.25 0.32 3.25 0,290 c old 11.3 3.4 0.30 d new 9.3 — 3.40 0.300 1 a 11.9 — 2 b 12.7 4.0 0.31 4.00 0.310 c old 11.5 — d new 15.1 — 31] RATE OF REGEXERATION—ZELEXV TABLE 20 (Continued) Series Re- gener- ation Length Length regen- erated Specific length regen- erated .•\veragc length regen- erated Average specific length , regen- erated 1 a 10.2 3.6 0.35 VI b 12.9 3.6 0.28 3.60 0.315 2 c old 11.2 3.1 0.28 d new 9.9 3.1 0.31 3.10 0.295 1 a 11.9 3.8 0.32 VII b 8.4 3.25 0.39 3.52 0.355 2 c old — d new 9.2 3.6 0.39 3.60 0.390 1 a 10.0 3.25 0.32 VIII b 8.5 3.4 0.40 3.32 0.360 2 c old 13.2 — d new 11.7 4.9 0.42 4.90 0.420 1 a 10.1 3.2 0.32 b 10.7 3.4 0.32 3.30 0.320 IX 2 c old 9.6 3.5 0.36 d new 8.7 3.0 0.34 3.25 0.350 1 a 12.1 3.6 0.30 b 10.1 2.3 0.23 2.95 0.265 X 2 c old — d new — 1 a 10.2 3.5 0.34 b 9.4 2.5 0.27 3.00 0.305 XI 2 c old 9.2 2.3 0.25 d new 8.9 2.5 0.28 2.40 0.265 1 a 11.0 — b 10.4 2.7 0.26 2.70 0.260 XII 2 c old 8.6 3.5 0.41 d new 8.7 3.5 0.40 3.50 0.405 Average 1 10.6 3.06 0.298 2 10.9 3.42 0.323 32 ILLINOIS BIOLOGICAL MONOGRAPHS [32 Experiment II Rana clamitans Series 3676-3765 Material and Method Ninety tadpoles with an average total length of about 40 mm. and an average tail length of 27 mm. were used in the experiment. The plan consisted in the removal of a portion of the tail in a part, S, of the individuals, the remaining part, F, being left un- injured at the time. After S had been regenerating a new tail for twenty-two days both S and P were operated upon. In S the regen- erating tails were removed by a cut which came at the border line be- tween the old and the new tissues. In P an operation was made similar to the original one on S and leaving the same amount of old tail in both S and P. The procedure is similar to that shown in Pigure 1. S and P were now allowed to regenerate and a direct comparison is possible between a second regeneration in S and a first regeneration in P. Measurements were made of regenerated lengths at 4, 6, 8, 10, 12^, 18 and 56 days. The operations were made at six levels corresponding approximately to the removal respectively of ^/jg, ^/iqi %> %> V2 and % of the tail. Pour of these levels, ^/jq) %, V2 ^^^ 73; had at least five individuals each for each regeneration. The other two levels, ^/jg and Vfe. tad less than five individuals per regeneration but are included in the tables though their averages are not as reliable as those of the others. The method as described agrees in principle with that pursued in Experiment I. It has a decided advantage over a direct comparison within a single individual because it eliminates the age factor as well as the effects of change in external conditions such as temperature and food. Data The results of the experiment are given in Tables 21 to 30 and. in Pigures 2 and 3. The data show on the whole a tendency for the second regeneration to remain in advance of the first for eight or ten days after the operation. The first regeneration then catches up and even slightly surpasses the other; this is apparent both when the regenerated lengths are taken directly and when they are corrected for difference in level of the cut and put in terms of specific regenerated length or the length regenerated per unit of removed length. In making the comparisons certain general features must be borne in mind. The maximum rate of regeneration is reached on or near the seventh day, earlier for the smaller removals and later for the larger removals. The whole regeneration, in so far as it is completed, is finished in nearly all cases at I2I/2 days, again somewhat earlier for the smaller and somewhat later for the larger removals. In the tadpoles used in the present experiment about four-tenths in length of the re- moved tail is replaced before regeneration stops. This was found to be 33] RATE OF REGENERATION— ZELENY 33 generally true of tadpoles of this size in Rana clamitans. The percent regenerated is somewhat greater for the smallest removals than for the others. After the maximum is reached there is a tendency toward de- crease of the regenerated region though this is hard to determine with accuracy because the boundarj- between old and new tissue becomes more and more obscure as time goes on. For this reason the data for 56 days of regeneration are not as reliable as the others. 4 6 8 10 121/2 18 — >- Days Figure 2. Specific regenerated lengths during the regenerative period for both first and second regenerations. Tadpole tail of Rana clamitans. Series 3676-3765. Broken line = second regeneration. Unbroken line = first regeneration. 2 5 7 9 ll',4 15% — >- Days Figure 3. Change in specific rate of regeneration during the regenerative period for both first and second regenerations. Tadpole tail of Rana clamitans. Series 3G76-3765. Unbroken line = first regeneration. Broken line = second regeneration. 34 ILLINOIS BIOLOGICAL MONOGRAPHS [34 At the four-day period the amount of regeneration is so small that there is a large probable error and these data should be used with cau- tion. For the ^/jg and % removals the number of individuals is so small that the data for these levels do not compare in accviracj' with the others and they will therefore be passed over for the present. The data are presented in Tables 21 to 30. Tables 21 to 26 give respectively the regenerations for tlie six different levels beginning with the shortest removal. Table 27 collects all the data of amounts regen- erated and Table 28 aU the data of specific amounts regenerated. Figure 2 gives in graphic form the specific amounts regenerated for each regeneration. Table 29 gives the differences between the first and second regenerations for each of the different levels at each of the seven times of measurement. It includes the differences in specific length as well as those in absolute length. The specific lengths furnish the better basis for comparison and will be used in the following discus- sion unless otherwise stated. Table 30 compares the specific rates in the first and second regenerations and Figure 3 gives the results in graphic form. Taking ujj the regeneration from the different levels and leaving out of consideration for the present the two levels with too small a number of individuals, the data for the ^/^ level as given in Table 4 are the first to be considered. There are five individuals for first and seven for second regenerations. The second regeneration is ahead in specific length from the fourth to the tenth daj'. At 12i/o days the two are tied and at .56 days the first is ahead. Eegeneration is completed in 121/2 days and beyond this time there is a decrease in regenerated ma- terial. The decrease is greater in the second than in the first regenera- tion, hence the ascendency of the latter at 56 days. During the Avhole period of active regeneration the second regeneration remains ahead. There are eight individuals for the first regeneration and eleven for the second at the % level (Table 24). The specific amounts of regenera- tion are strikingly similar throughout the whole period of regeneration. The two departures from equality are an advantage of 0.01 for the second regeneration at 8 daj'S and a disadvantage of 0.02 at IS days. These ■departures are in the direction of the general rule observed at other levels that the second regeneration tends to be ahead at the earlier pe- riods and the first at later periods, the advantage in the later case being due to the earlier completion of regeneration and absorption of regener- ated material in the second regenerations than in the first ones. In this instance the first regeneration does not gain an advantage until after the second has reached its maximum. At the 1/2 level there are 5 individiials for the first regeneration and 8 for the second (Table 25). The second is ahead until the eighth day. 35] RATE OF REGENERATION— ZELEXY 35 Begiuniug with the tenth day the first is ahead. In general the advan- tage of the first increases as time goes on. The growth of new tissue does not terminate until the eighteenth day or after. At the ^3 level there are five individuals for the first and tea for the second regeneration (Table 26). The second is ahead of the first until the tenth day, after which the first is in the lead. Regeneration is not stopped until the eighteenth day or later. At all four of these levels the specific length of the second regen- eration tends to be ahead until the tenth day (Table 28 and Figure 2). The maximum rate of regeneration is reached before this time and some- what earlier by the second than by the first regeneration, hence the relative gain by the latter after the tenth day (Table 30 and Figure 3). The stopping of regeneration also comes earlier for the second than for the first regeneration as does the beginning of absorption of regenerated material. The data in Experiment I concern the amount of regeneration at six and at eight daj's. At the corresponding times in Experiment II the second regeneration is ahead of the first. There is a full agreement between the two experiments in this regard. The more rapid rate of the second regeneration at the start may at first sight seem to be due to the presence of at least some cells which have been actively engaged in previous regenerations. If the second cut comes outside of the boundary between old and new cells the latter cover the whole new cut surface. Even if the cut seems to be exactly at the original cut level there will be some new cells at the regenerating surface. These cells which are already regenerating may be expected to adjust themselves more readily to the new conditions than old ones which have not been engaged in such a process. In another place the relative rates from old and from new tissue are described and a .slight early difference favoring the new tissue is made out. While tliis slight initial advantage may be explained in this way it is probably confined to the period of cell migration and is not a factor in the period of cell di\'ision which begins on the second day or later. It is evident that on the wliole the control of rate is not a matter inherent in the cells in the neighborliood of the cut surface. Indications point rather to a more central control of the process. 36 ILLINOIS BIOLOGICAL MOXOGRAPHS 1 36 TABLE 21 Rana clamitans Series 3676-3765 Comparison of first and second regenerations Age factor eliminated One-eighteenth of tail removed Catalog number Re- moved lengtb 1.4 1.7 1.5 1.3 1.6 1.6 1.6 1.5 Length regenerated in mm. 4 Days 0.24 0.30 6 Days 8 Days 10 Days 1.0 0.8 m Days 18 Days 56 Days First 3706 3742 0.54 0.40 0.9 0.7 1.0 0.9 0.9 0.9 0.7 0.7 eration Average Second regen- eration 3676 3682 3730 3754 0.27 0.18 0.39 0.06 0.60 0.60 0.75 0.55 0.9 0.9 0.9 0.9 1.0 1.0 0.9 1.1 1.0 1.0 0.9 1.2 1.0 1.0 0.9 1.2 0.7 1.1 0.7 1.1 Average Av. length— -First regen. 0.27 0.47 0.8 0.9 0.9 0.9 0.7 Av. length- -Second regen. 0.22 0.62 0.9 1.0 +0.1 1.0 1.0 0.9 Increase or decrease —0.05 -fO.15 +0.1 +0.1 +0.1 +0.2 Specific Ig.- —First regen. 0.17 0.30 0.53 0.58 0.67 0.61 0.60 0.67 40.07 0.45 Speciflc Ig.- -Second regen. 0.15 0.42 +0.12 0.60 0.67 0.60 Increase or decrease —0.02 +0.07 +0.09 4 0.06 4 0.15 37] RATE OF REGENERATION — ZELEWy TABLE 22 Rana clamitans Series 3676-3765 Comparison of first and second regenerations Age factor eliminated One-tenth of tail removed Catalog number Re- moved length mm. 2.5 Length regenerated in mm. 4 Days 6 Days 8 Days 10 Days 12J Days 18 Days 56 Days 3688 0.12 0.3 0.3 0.7 0.9 0.9 0.7 3707 3.2 0.24 0.8 1.1 1.4 1.4 1.3 0.7 First 3724 2.6 0.06 0.5 o.s 1.1 1.4 1.4 1.2 regen- 3743 2.5 0.03 0.1 0.4 0.8 1.0 1.0 1.7 eration 3760 3.1 2.6 0.30 0.6 0.9 1.1 1.2 1.1 1.1 Average 3677 2.0 0.30 0.6 0.9 0.9 0.9 0.9 0.7 3696 2.1 0.48 0.8 1.0 1.1 1.0 1.0 0.7 3713 2.8 0.36 0.8 0.9 0.9 0.9 0.9 0.5 Second 3719 3.1 0.36 0.8 1.1 1.4 1.4 1.3 — regen- 3749 2.8 0.30 0.6 1.2 1.4 1.4 1.3 0.9 eration 3750 3.5 0.48 1.3 1.7 1.9 2.0 2.0 — 3701 3.2 2.8 0.42 0.8 1.1 1.2 1.3 1.3 — Average Av. length- -First regen. 0.15 0.5 0.7 1.0 1.2 1.1 1.1 Av. length- -Second regen. 0.39 0.8 1.1 +0.4 1.3 1.3 +0.1 1.2 0.8 Increase or decrease -fO.24 +0.3 +0.3 +0.1 —0.3 Specific Ig.- -First regen. 0.06 0.14 0.18 0.27 0.39 -fO.IZ 0.38 0.46 0.42 0.42 Specific Ig.- —Second regen. , 0.30 +0.12 0.46 0.46 0.43 0.29 Increase or decrease ^ 0.08 i 0.08 0.00 +0.01 —0.13 38 ILLINOIS BIOLOGICAL MONOGRAPHS [38 TABLE 23 Rana elamitans Series 3676-3765 Comparison of first and second regeneration Age factor eliminated One-sixth of tail removed Catalog number Re- moved length 5.3 Length regenerated in mm. 4 Days 0.54 6 Days 1.2 8 Days 10 Days 12i Days 18 Days S6 Days 3708 1.9 2.1 2.3 2.3 1.8 First 3726 4.3 0.42 0.9 1.3 1.4 1.4 1.4 1.4 regen- 3762 4.1 0.57 1.0 1.2 1.5 1.7 1.8 1.4 eration Average 4.6 3678 5.0 0.20 0.5 0.9 1.0 1.1 1.1 3684 5.5 0.15 0.7 1.2 1.4 1.4 1.5 1.4 Second 3702 4.7 0.42 0.8 1.1 1.3 1.3 1.3 1.3 regen- 3720 4.6 0.06 0.3 1.3 1.8 2.3 1.9 2.0 eration 3756 4.8 4.9 0.36 0.51 1.0 1.3 1.7 1.8 1.6 — Average Av. length— -First regen. 1.0 1.5 1.7 1.8 1.6 1.8 1.5 Av. length — Second regen. 0.24 0.7 1.2 1.4 1.5 1.6 Increase or decrease —0.27 —0.3 —0.3 0.33 —0.3 —0.2 —0.3 +0.1 Specific Ig — -First regen. 0.11 0.05 0.22 0.37 0.29 0.39 0.39 0.34 Specific Ig — -Second regen. 0.14 0.24 0.33 0.31 0.33 Increase or decrease —0.06 —0.08 —0.09 —0.08 —0.06 —0.08 —0.01 39] R.-^ITE OF REGEXERATIOX — ZELEXY 39 TABLE 24 Rana clamitaiis Series 3676-3765 Compai'ison of first and second regenerations Age factor eliminated One-tliird of tail removed Catalog number Re- moved length 1. ngth regenerated in mm. 4 Days 6 Days S Days 10 Days 12} Days 18 Days 56 Days 3690 9.7 0.48 1.0 1.7 2.4 2.6 2.7 2.2 3709 8.8 0.48 1.3 2.0 2.6 3.2 3.4 3.3 3727 8.3 0.48 1.1 1.6 2.0 2.2 2.2 2.2 First 3745 10.0 0.54 1.8 2.4 3.8 4.4 4.8 4.2 regen- 3744 6.0 0.36 1.0 1.3 1.7 1.8 1.7 — eration 3761 6.6 0.39 1.0 1..-. 1.9 2.2 2.3 1.8 3763 8.5 0.57 1.1 l.S 2.4 2.9 3.1 — 3689 6.3 0.30 0.30 0.7 1.2 1.5 1.6 1.7 1.4 Average 8.2 3679 8.4 0.7 1.4 1,9 1.9 2.1 — 3685 9.3 0.60 1.2 1.9 2.3 2.8 3.0 2.6 3697 7.3 0.48 1.2 1.7 2.2 2.4 2.3 2.1 3703 9.3 0.45 1.3 2,0 2.5 2.6 2.5 2.5 3715 7.9 0.24 0.9 1.7 2.3 2.6 2.6 2.8 Second 3721 8.7 0.57 1.3 1.9 2.3 2.6 2.3 2.2 regen- 3733 8.5 0.36 1.0 1.9 2.4 2.6 2.6 2.8 eration 3734 8.5 0.48 2.1 3.1 4.5 5.7 6.4 6.6 3739 9.6 0.36 1.0 l.S 2.4 3.0 2.9 — 3751 6.7 0.45 1.1 1,7 2.1 2.4 2.5 2.3 3757 8.0 8.4 0.42 1.2 2.1 2.8 3.1 3.2 3.1 Average 1.1 Av. length- -First reg en. 0.45 1.7 2.3 2.6 2.7 2.5 Av. length- -Second re gen. 0.42 1.1 1.8 2.3 2.6 2.6 2.5 Increase or decrease —0.03 0.0 +0.1 0.0 0.0 —0.1 0.0 Specific Ig.- —First reg en. 0.05 0.13 0.21 0.28 0.31 0.33 0.30 Specific Ig.- —Second r egen. 0.05 0.13 0.22 0.28 0.31 0.31 0.30 Increase or decrease 0.00 0.00 +0.01 0.00 0.00 —0.02 0.00 40 ILLIXOIS BIOLOGICAL MOXOGKAPHS [40 TABLE 25 Rana clamitans Series 3676-3765 Comparison of first and second regenerations Age factor eliminated One-half of tail removed Catalog number Re- moved length mm. Length regenerated in mm. 4 Days 6 Days 8 Days 10 Days m Days 18 Days 56 Days 3710 12.3 0.42 1.8 2.9 3.7 3.9 3.9 3.9 3728 12.8 0.60 1.7 2.8 3.9 4.8 5.4 5.8 First 3746 13.3 0.54 1.7 2.4 4.1 5.7 7.0 6.8 regen- 3764 14.6 0.42 1.3 2.5 4.2 5.3 6.8 6.5 eration 3765 12.2 0.30 1.5 2.3 3.2 3.9 4.5 4.5 Average 13.0 3686 14.5 0.60 2.1 3.4 4.8 5.3 5.2 — 3698 14.9 0.50 1.5 3.3 4.3 5.0 5.4 5.4 3704 14.5 0.45 2.2 3.3 4.4 5.2 5.5 5.4 Second 3716 12.7 0.39 1.7 2.4 3.4 4.2 5.1 4.4 regen- 3722 12.5 0.60 1.6 2.6 3.6 3.9 3.5 4.2 eration 3740 13.9 0.30 1.1 2.1 3.0 4.6 5.6 6.8 3752 12.2 0.54 1.7 2.5 3.4 4.1 4.0 — 3758 11.0 13.1 0.60 1.5 2.2 2.9 3.6 4.1 4.9 Average Av. length— -First regen. 0.46 1.6 1.7 2.6 3.8 4.7 5.5 5.5 Av. length- -Second regen. 0.50 2.7 3.7 4.4 4.8 —0.7 5.2 Increase or decrease -f0.04 +0.1 + 0.1 —0.1 —0.3 —0.3 Specific Ig.- -First regen. 0.03 0.12 0.20 0.29 0.36 0.42 0.42 Speciflc Ig- -Second regen. 0.04 0.13 +0.01 0.21 0.28 0.34 0.37 0.40 Increase or decrease +0.01 +0.01 -0.01 —0.02 -^0.05 —0.02 41] RATE OF REGESERATION — ZELEXY TABLE 26 Comparison of first and second regenerations Age factor eliminated Rana clamitans Series 3676-3765 Two-thirds of tail removed Catalog number Re- moved length mm. 16.8 Length regenerated in mm. 4 Days 6 Days 8 Days 10 Days 12i Days 18 Days 56 Days 3692 0.51 1.1 2.2 3.2 4.3 5.0 5.2 3693 17.2 0.48 1.8 3.3 5.0 6.5 7.3 6.6 First 3711 17.0 0.54 1.8 3.6 5.6 7.0 7.7 8.3 regen- 3729 16.1 0.48 1.9 3.3 4.6 5.5 6.7 6.4 eration 3749 16.2 16.7 16.0 0.54 1.2 2.7 4.2 5.6 7.1 7.8 Average 3680 0.60 1.9 3.0 4.2 5.2 6.4 6.6 3681 21.2 0.84 3.0 4.0 5.6 6.3 7.3 7.2 3687 19.7 0.54 3.6 5.6 6.0 6.6 7.0 — 3699 21.0 0.54 2.2 4.3 5.9 7.1 7.5 7.2 Second 3705 17.6 0.72 2.0 3.6 4.8 6.4 6.2 6.0 regen- 3717 17.6 0.42 2.6 3.6 5.2 6.0 6.7 6.4 eration 3723 18.4 0.30 2.3 3.7 5.3 6.5 8.1 8.3 3735 16.5 0.48 2.0 3.4 5.5 6.5 7.8 8.0 3741 16.0 0.30 1.9 3.0 4.4 5.8 6.9 7.0 3753 16.8 18.1 0.42 2.0 2.5 3.8 5.2 6.4 7.1 Average 0.51 Av. length- -First regen. 1.56 3.02 4.52 5.78 6.76 6.86 Av. length- -Second regen. 0.52 2.35 3.67 5.07 6.16 7.03 7.09 Increase or decrease -rO.01 +0.79 +0.65 +0.55 +0.38 +0.27 + 0.23 Specific Ig.- —First regen. 0.03 0.03 0.09 0.13 0.18 0.20 0.27 0.35 0.40 0.41 Specific Ig.- -Second regen. 0.28 0.34 0.39 0.39 Increase or decrease 0.00 +0.04 +0.02 +0.01 —0.01 —0.01 —0.02 42 ILLIXOIS BIOLOGICAL MOXOGRAPHS TABLE 27 Rana clamitans Series 3676-3765 Comparison of first and second regenerations Age factor eliminated Average lengtlis regenerated in mm. Approx. fraction of tail removed Re- gener- ation Number of individ- uals Average length removed .\verage length regenerated 4 Days 6 Days 8 Days 1 i 0.8 0.9 0.7 10 Days 12i Days 18 Days 56 Days '/,« 1 2 1.5 0.3 0.5 0.9 0.9 0.9 1.0 0.7 2 4 1.5 0.2 0.6 1.0 1.0 .0.9 '^u. 1 5 2.6 0.1 0.5 1.0 1.2 1.1 1.1 2 7 2.8 0.4 0.8 1.1 1.5 1.3 1.3 1.2 0.8 V„ 1 3 4.6 0.5 1.0 1.7 1.8 1.8 1.5 1.5 2 5 4.9 0.2 0.7 1.2 1.7 1.4 1.6 1.6 v.. 1 8 8.2 0.4 1.1 2.3 2.6 2.7 2.5 2 10 8.4 0.4 1.1 1.8 2.6 2.7 2.3 2.6 2.6 2.5 V, 1 5 13.0 0.5 0.5 1.6 3.8 4.7 5.5 4.8 5.5 2 8 13.1 1.7 3.7 4.4 5.2 1 5 16.7 0.5 0.5 1.6 3.0 4.5 5.8 6.8 7.0 6.9 '■■■• 2 10 18.1 2.3 3.7 5.1 6.2 7.1 43] RATE OF RECENERATIOX—ZELE.XY 43 TABLE 2S Rana clamitans Series 3676-3765 Comparison of first and second regenerations Age factor eliminated Specific lengths regenerated . Approx. fraction of tail removed Re- gener- ation Number of individ- uals Average length -emoved in mm. Specif c length regene rated in mm. 4 Days 6 Days 8 Days 0.53 0.60 0.27 0.39 0.33 0.24 0.21 0.22 0.20 0.21 0.18 0.20 10 Days 12J Days 18 Days 56 Days •As 1 2 1.5 0.17 0.30 0.42 0.58 0.61 0.60 0.45 2 4 1.5 0.15 0.67 0.67 0.67 0.60 '/,o 1 5 2.6 0.06 0.18 0.30 0.38 0.46 0.42 0.42 2 7 2.8 0.14 0.46 0.46 0.43 0.29 •/« 1 3 4.6 0.11 0.22 0.14 0.37 0.39 0.39 0.34 2 5 4.9 0.05 0.29 0.33 0.31 0.33 '/a 1 8 8.2 0.05 0.13 0.13 0.28 0.31 0.33 0.30 2 10 8.4 0.05 0.28 0.31 0.31 0.30 •■/, 1 5 13.0 0.03 0.12 0.13 0.29 0.36 0.42 0.42 2 8 13.1 0.04 0.28 0.34 0.37 0.40 % 1 5 16.7 0.03 0.09 0.13 0.27 0.35 0.40 0.41 2 10 18.1 0.03 0.075 0.077 0.28 0.34 0.413 0.408 0.005 0.39 0.427 0.413 0.014 0.39 All levels- Averac J e— First 0.173 0.208 0.287 0.310 0.362 0.390 All levels- Avera je — Seco id 0.377 0.385 First ahead - - 0.023 0.015 0.005 Second ahead 0.002 0.035 - - 44 ILLINOIS BIOLOGICAL MONOGRAPHS [44 \ s cc •a u M < ro (D -^ H d ' ' ^ !C -*-^ a a o 3 ^; iM I— 1 to 1-t o o CO o cq O io "' O o o <31 o o o n^ + 4- . 1 1 + 1 1 1 + 1 t- rt 00 in ^ > y~\ o »H c- o o 2 "> o o O o o o O + + + + 1 1 1 1 1 1 + 1 oo o to ^ ^ T-l 1-H o o CO o o o o o 1 o o o d^' o + + + 1 1 1 1 ! 1 + 1 -Ji oo oo Q ^ T^ PO CO o iH o CO o o o o o o <^ o + + + + 1 1 1 1 + + CO ^ ^ t- ■<*< ?3 CO a> 1-1 o ^ o Ir- C^l Q o o o o o o <3; + + + 1 1 + + + + + + r M CM oo ^ w o o O o o o o o c> o (~, + + + + 1 1 + + + + tH o CO § CO o ^ o o o ^ o o o o o o o o o 1 1 + + 1 1 + I a Ml a a a FJ ja a 5 a ^ n ■a a a a a d a be d d a 60 d ■2 a JH d ai d a d 0) /3 Id J3 o !d ja !d J3 J3 td ja o W) 3 M M u W) ■q Ml W) Q a d d (U d d d p< Pi 0) di lU c. a kJ cc J cc 1-1 CC J M ►J m hJ M "c c ° -^ « >- -* ^ lO ° 00 o E •; w lO « « ^ u. 45] RATE OF REGEXERATIOS — ZELENY 4S Rana clamitans Series 3676-3765 Specific rates of first and second regenerations during each of tlie time periods Approx. Re- gener- ation Num- Der of indi- viduals length removed Specific rate of regeneration of Uil removed 0-4 Days 4-6 Days 6-8 Days 8-10 Days 10-12* Days 121-18 Days 18-56 Days ^/.s 1 2 2 1.5 0.042 0.065 0.115 0.025 0.015 —0.002 —0.004 4 1.5 0.037 0.135 0.090 0.035 0.000 0.000 —0.002 >/xo 1 2 5 2.6 0.015 0.040 0.045 0.055 0.040 —0.007 —0.001 7 2.8 0.035 0.080 0.045 0.035 0.000 —0.005 —0.004 "/« 1 2 3 4.6 0.027 0.055 0.055 0.020 0.010 0.000 —0.001 5 4.9 0.012 0.045 0.050 0.025 0.025 —0.004 0.001 'U 1 2 8 8.2 0.012 0.040 0.040 0.035 0.015 0.004 —0.001 10 8.4 0.012 0.040 0.045 0.030 0.015 0.000 —0.000 '/. 1 2 5 13.0 0.007 0.045 0.040 0.045 0.035 0.011 0.000 S 13.1 0.010 0.045 0.040 0.035 0.030 0.005 0.001 =/3 1 2 5 16.7 0.007 0.030 0.045 0.045 0.040 0.009 -1-0.000 10 IS.l 0.007 0.050 0.035 0.040 0.030 0.009 0.000 All levels — Average — First 0.018 0.046 0.057 0.037 0.026 0.002 —0.001 All levels — Average — Second 0.019 0.066 0.051 0.033 0.017 0.001 —0.001 First ahead 0.006 0.004 0.009 0.001 Second ahead 0.001 0.020 46 ILLINOIS BIOLOGICAL MOXOGRAPHS [46 Experiment III Amblystoha punctatum Series 3962-3999 Material and Method Eggs of Amblystoma puoctatum in the cleav- age stages were collected on Marcli 18, 1913, and hatched in the labo- ratory on April 9. The first operations were made on April 23, at which time also five controls were killed and preserved. These when measured gave an average total length of 13.1 mm. and an average tail length of 5.3 mm. Ninety individuals were used for the regeneration study. In thirty individuals two-thirds in length of the tail was re- moved on April 23. The regenerated portion in these was removed on May 10 and at the same time in a second thirty individuals two- thirds of the tail was removed. On May 21 the first thirty were operated on for the third time, the second thirty for the second time, and the third thirty for the first time. To insure as accurate a compari- son as possible the ninet.y individuals though they were approximately of equal size were divided into thirty groups of three each, a selection being made so that the three members of a group were as much alike as possible. In each group one of the tliree members was used for the first regeneration, one for tlie second and the third for the third regen- eration. This procedure gave a possibility of comparing the first, second and third regenerations without error due to difference in size, age, or in external conditions. Three individuals from each thirtj- were killed two days after the last operations, four in four days, five in six days, five in eight days, six in ten da.ys and seven in fourteen days. At the end of the experiment, control individuals gave an average total lengtli of 31.5 mm. and an average tail lengtli of 10.5 mm. Data The data are given in Tables 31 and 32. The specific amounts of regeneration were not determined because the removed lengths were alike and hence the comparison of absolute lengths gives the same re- sults as a comparison of specific amounts. The average regenerated lengths at each of the six different times will be taken up first. At two days the average regenerated lengths for the first, second and third regenerations are respectively 0.22, 0.25 and 0.26 mm. At four days the corresponding amounts are 0.66, 0.75 and 1.00. At six days they are 1.36, 1.40 and 1.36, but the low value of the third regeneration is due to a single exceptional individual. At eight days the figures are 2.18, 2.68 and 2.68. At ten days they are 3.55, 3.82 and 4.20 and at fourteen days 5.34, 6.12 and 6.08. In all cases, except the one at six days explained above, both second and third regenerations are ahead of the first. Tlie tliird regeneration is greater than the second at two, four and ten days, is equal to the second at eight days and less than the second at six and fourteen days. Since the low 47] RATE OF REGENERATION —ZELENY A7 average for the third regeneration at six days is due to a single excep- tional individual it is more proper to put the third ahead of the second at this time. A comparison of the three regenerations by individual cases is shown in Table 32. At each of the six times taken the number of cases showing a more rapid regeneration is greater for the third regeneration than for the first and also greater for the second than for the first. The third is ahead of the second at two times (more properly three times) and equal to the third at four times (more properly three). When all the individual cases are taken together both third and second regenerations are again distinctly ahead of the first as showni by the totals in Table 32. The third is ahead of the second in twelve cases (more properly thirteen) and the second ahead of the first in eight cases (more properly seven). Each of the three compai-isons shows that both second and third regenerations are more rapid than first regenerations. The third regen- eration shows a slight advantage over the second instance in all three of the comparisons. In this instance the difference can not be diie to the presence of newly regenerated cells in the one case and not in the other. TABLE 31 Amblystoma punctatum Series 3967-3998 Comparison of lengths of first, second and third regenerations Age factor eliminated Regener- ation time in days Catalog number 2 3967 3968 3969 Average 4 3970 3971 3972 3973 Average Regenerated lengths ir mm. First regeneration Second regeneration Third regeneration 0.2 0.25 0.2 0.25 0.3 0.2 0.3 0.27 0.2 0.22 0.25 0.26 0.75 0.7 0.5 0.7 0.9 0.75 0.6 1.0 1.6 0.8 0.6 0.66 0.75 1.00 48 ILLIXOIS BIOLOGICAL MONOGRAPHS 148 TABLE 31 (Continued) Amblystoma punctatum Series 3967-3998 Comparison of lengths of first, second and tliird regenerations Age factor eliminated Regener- Regenerated lengths in mm. ation time Catalog in number First Second Third days regeneration regeneration regeneration 3974 1.2 1.2 3975 1.4 1.5 1.5 3976 1.3 1.4 1.6 6 3977 1.5 1.2 0.6 3978 1.4 1.7 1.7 Average 1.36 1.40 1.36 3980 1.7 2.4 2.7 3981 1.9 — 2.9 3982 2.3 2.6 3.0 8 3984 2.4 2.8 2.1 3985 2.6 2.9 2.7 Average 2.18 2.68 2.68 3986 4.1 3.8 4.7 3987 3.6 3.7 — 3988 3.5 — 3.9 3989 2.6 3.9 3.5 10 3990 3.2 4.25 4.6 3991 4.3 3.5 4.35 Average 3.55 3.82 4.20 3992 5.5 5.5 5.7 3993 5.0 5.75 5.7 3994 — 6.7 6.9 3995 5.0 5.7 6.9 14 3997 6.9 6.7 5.2 3998 4.3 6.35 — Average 5.34 6.12 6.08 49] RATE OF REGENERATION— ZELENY TABLE 32 Amblystoma punctatum Series 3967-399S Age factor eliminated Comparison of lengtlis of first, second and third regenerations Comparison of individual cases Comparisons 3rd regen. > 1st 3rd regen. = 1st 3rd regen. < 1st 2nd regen. > 1st 2nd regen. = 1st 2nd regen. < 1st 3rd regen. > 2nd 3rd regen. = 2nd 3rd regen. < 2nd Two days Four days Six Eight days days 3 4 1 1 3 4 1 1 1 2 2 1 2 Ten days Fourteen days Totals 20 1 4 ExPERiJiENT IV Amblystoma punctatum Series 3962-3999 The series used for Experiment III furnishes another set of data for the effect of successive removal. When the third operation was made the removed regenerated tails of the first thirty individuals represented an eleven-day second regeneration and those of the second thirty-indi- viduals an eleven-day first regeueration. A direct comparison is thus possible between the first and the second regenerations. It is not possi- ble to make a cut exactly at the border line between old and new tissue and therefore the measurement of the removed regenerating tail is not as accurate a determination as is the direct measurement of a regener- ating unremoved tail. The data are shown in Table 33. Twenty-five individuals are avail- able for each regeneration. The average of the first regenerations is 4.55 =0.11 and of the second regenerations 4.50 -0.10. The first regen- eration is ahead of the second in ten cases, the second is ahead of the first in twelve cases and three cases are equal. The first comparison shows a slight difference in favor of the first regeneration but this is so much less than the probable error that it can not be considered as significant. The second comparison shows a slight advantage in favor of the second regeneration. On the whole the data indicate essential equality between the first and the second regenerations at eleven days. 50 ILLIXOIS BIOLOGICAL MONOGRAPHS TABLE 33 Amblystoma punctatum Series 3962-3999 Age factor eliminated Comparison of first and second regenerations Eleven days [50 First Second First Second First Catalog regen. regen. ahead ahead and second number mm. mm. of second of first equal 3967 4.0 4.4 0.4 3968 3.7 3.5 0.2 3969 4.9 5.1 0.2 3970 4.5 4.7 0.2 3972 4.7 4.7 * 3973 3.9 4.3 0.4 3975 4.5 5.7 1.2 3976 4.9 4.9 * 3977 3.8 3.7 0.1 3978 4.1 4.5 0.4 3980 4.9 5.0 0.1 3981 3.5 4.4 1.1 3982 5.0 4.7 0.3 3984 5.1 4.0 1.1 3985 5.8 4.3 1.5 3986 3.8 4.1 0.3 3989 4.8 5.5 0.7 3990 5.5 4.3 1.2 3991 4.1 4.6 0.5 3992 4.6 4.1 0.5 3993 4.5 4.5 * 3994 4.9 5.0 0.1 3995 4.5 4.0 0.5 3997 5.1 4.2 0.9 3998 4.8 4.3 0.5 4.55±0.11 4.50+0.10 ten times twelve times three times Experiment V Amblystoma punctatum Series 6042-6100F This series was devised for a study of the etfect of repeated removal of the tail upon the rate of metamorphosis. The removed tails were preserved and they give some data on the comparison of successive regenerations. The interest of the results lies in the fact that the suc- cessive regenerations are compared within single individuals. Thus the effect of the age factor is not eliminated. Environmental differences such as those of temperature may also be factors. The eggs were hatched on March 25 to 29. 1915. Approxi- matelv one-half in length of the tail was I'emoved in each of the indi- 51] RATE OF REGENERATION— ZELENY 51 viduals ou April 5. The new tissue was removed on April 17 and again ou May 1, May 10 and May 19, making five removals in all. The second removal gives the fii-st regeneration, the third the second, and so on. The regenerated lengths were therefore determined by measurement of removed parts. This does not give as accurate a deter- mination as does direct measurement without removal because the cut can not in ordinary practice be made exactly at the border line between old and new tissue. The data are given in Table 34. The first regeneration covers a twelve-day period, the second fourteen days and the third and fourth each nine days. The third and fourth regenerations are the only ones that have the same time interval. Ten individuals are available for this comparison. The average for the third regeneration for these ten is 1.30 mm. and of the fourth regeneration 1.17 mm. When all individuals are taken without regard to representation of both regenerations the average for the third regeneration is 1.28 and for the fourth 1.17. In seven of the ten former cases the third is ahead of the fourth regeneration, in two they are tied and in one the fourth is ahead of the third. The data therefore show an advantage of the third over the fourth regeneration. The first regeneration ran twelve days and the second fourteen days. The maximum rate of regeneration comes on or near the ninth day and the rate has declined to a low point by the fourteenth day. However it is not possible to make the necessary correction because of lack of data on the rate curve for this particular set of larvae. Some facts may however be obtained by a comparison. Sixteen individuals for each of the two regenerations are available for comparison. The average for the first regeneration in these is 2.06 mm. and for the sec- ond 2.01 mm. In seven the first is ahead of the second, in seven the second is ahead of the first, and two are tied. When all individuals are taken without regard to representation of both regenerations the average for the first regeneration is 1.99 -0.03 mm. for a twelve-day period and for the second regeneration 2.01 for a foiirteen-day period. The difference between the two values is not significant, but when the longer time interval taken by the second regeneration is considered the conclusion is reached that the first regeneration is more rapid than the second. The data thus indicate a progressive decrease in rate from the first to the fourtli regenerations. This result taken in connection with the results obtained from the experiments in which the age factor is eliminated makes it highly probable that the decrease in rate of regen- eration observed here is due to increase in age and not to the effect of successive removal. 52 ILLINOIS BIOLOGICAL MOXOGRAPHS Amblystoma punctatum Series 6042-6100 F Age factor eliminated Successive regenerations in single individuals First Second Third Fourth regeneration regeneration regeneration regeneration Catalog mm. mm. mm. mm. number Twelve Fourteen Nine Nine days days days days 6042 2.0 2.4 1.6 6043 1.8 2.5 1.5 1.4 6044 2.2 6046 2.2 2.2 1.3 1.3 6047 2.2 2.1 1.4 1.3 6048 2.0 2,2 6049 2.3 1.8 1.5 1.0 6050 1.8 6052 2.0 6053 1.9 6055 1.9 6056 1.7 6057 2.0 6058 2.0 6059 1.9 6061 1.5 6062 2.3 6065 1.5 6067 1.6 6068 1.6 6071 2.0 6072 1.9 6076 2.1 6077 2.0 6079 2.0 6080 1.8 6081 1.9 6082 1.9 2.0 1.0 1.1 6083 2.0 2.1 1.3 1.1 6084 1.9 6085 2.0 2.0 1.5 1.5 6086 2.2 6087 1.8 1.0 0.8 6088 2.1 2.4 1.4 1.2 6090 2.5 6093 2.3- 2.2 1.0 O.S 6094 2.1 53] RATE OF REGENERATION— ZELESY TABLE 34 (Continued) 53 First Second Tliird Fourth regeneration regeneration regeneration regeneration Catalog mm. ram. mm. mm. number Twelve Fourteen Nine Nine days days days days 6096 2.1 6097 2.5 1.6 6098 1.8 2.1 6099 2.2 6100D 2.0 6100E 1.8 1.6 6100F 2.2 2.0 1.1 1.0 Average 1.99±0.03 2.01 1.28 1.17 Rate per day 0.166 0.144 0.142 0.130 Experiment VI Bufo americ.vnus Series 6283-6323 This series was designed for the study of the effect of successive removal of the tail upon the rate of metamorphosis. The lengths of the removed regenerating tails however are of some value in a com- parison of successive regenerations though here as in Experiment V age and external factors are not eliminated. The eggs were laid on April 20-21, 1915. The tadpoles were col- lected on April 27 and the first removals were made on April 28. The first metamorphosis was completed on June 11. Tlie average total length at the time of the first removal was 10.9 mm. and the average tail length 6.4 mm. The average removed length was 3.8 mm., which is approximately 60 per cent of the tail length. Tlie second removal was made on May 7 and gives a nine-day period for tlie first regeneration. The third removal of May 17 gives a ten-day period for the second regeneration. The fourth removal on May 26 gives a nine-day period for the third regeneration. As in the case of Experi- ment V the cuts could not in practice be made to come exactly at the border line between old and new tissue and the accuracy of the meas- urements is therefore not as great as in those cases in which the lengths were taken directly from the animal without removal of the tail. The data are shown in Table 35. The first, second and third re- generation lengths are given for sixty individuals. The first and third regenerations have the same time interval and are therefore directly comparable. The average for the first regeneration is 1.94 "0.02 mm. 54 ILLINOIS BIOLOGICAL MONOGRAPHS IS4 and for the third 1.80 '0.03 mm., a difference in favor of the first re- generation of 0.14 *0.05 mm. This represents a regeneration of 0.51 mm. per unit of removed length in the first regeneration and 0.47 mm. per unit in the third regeneration. A comparison of individual cases shows that the first regeneration is ahead of the third in 36 individuals, the third is ahead of the first in IS individuals and 5 are tied. The differ- ence between the two regenerations is thiis probablj' significant. As in Experiment V the decrease is probably due to the age factor. The second regeneration has a time interval of ten days, one day more than the first and third regenerations. In the absence of know- ledge concerning the rate curve for toad tadpoles of this age no cor- rection can be applied. The rates per day for the three regenerations are however given in the table. TABLE 35 Bufo americanus Series 6283-6323 Age factor not eliminated Successive regenerations of tail First Second Third Catalog regeneration regeneration regeneration number Nine days Ten days Nine days Length in mm. Length in mm. Length in mm. 6283 a 2.1 2.0 1.9 b 1.5 2.1 1.6 c 1.7 2.1 1.7 6285 a 2.3 1.9 2.0 b 2.3 2.1 2.0 c 2.0 2.4 2.4 6287 a 1.9 2.1 2.1 b 2.1 2.0 1.9 c 1.8 2.3 1.8 6289 a 1.9 2,3 2.2 b 2.1 2.0 1.7 c 2.2 2.1 2.0 6291 a 2.1 2.2 2.0 b 1.9 1.9 1.8 c 2.0 1.9 1.4 6295 a 2.0 -'.0 2.U b 2.1 2.1 1.8 c 1.8 1.9 1.9 6297 a 1.9 2.0 2.0 b 1.9 2.1 1.8 1.9 2.0 1.8 6299 a 1.8 2.0 1.5 b 2.0 2.0 1.7 55] RATE OF REGENERATION— ZELENY TABLE 35 (Continued) First Second Third Catalog regeneration regeneration regeneration number Nine days Ten days Nine days Length in mm. Length in mm. Length in mm. c 1.8 1.9 1.3 6301 a 2.0 1.5 1.4 b l.S 1.9 1.2 c 1.7 1.8 1.4 6303 a 1.9 1.9 1.3 b 1.9 1.9 1.5 c 1.8 2.0 1.6 6305 a 1.9 2.0 1.5 b 1.7 1.8 2.0 c 2.1 2.2 1.9 6307 a 2.1 1.8 2.0 b 1.9 2.0 2.1 c l.S 1.9 1.8 6309 a 2.0 1.9 1.9 b 1.9 1.9 2.0 c 2.0 2.0 1.7 6311 a 1.7 2.4 1.8 b. 1.8 1.9 2.1 c 2.0 2.1 2.0 6313 a 2.0 2.3 1.7 b l.S 1.7 2.0 c 1.7 1.7 1.5 6315 a 1.9 2.4 2.0 b 2.0 2.1 1.6 c 1.9 1.9 6317 a 1.9 2.0 1.3 b l.S 2.2 2.0 c 1.9 2.1 1.8 6319 a 2.2 2.0 2.0 b 2.1 2.0 2.0 c 1.9 1.9 2.0 6321 a 2.3 2.0 l.S b 1.7 2.0 1.9 c 2.0 2.3 1.3 6323 a 2.1 2.1 1.7 b 1.8 2.3 2.0 c 1.9 2.2 2.0 Average 1.94±0.02 2.02=0.02 1.80±0.03 Rate per day 0.216 0.202 0.200 56 ILLINOIS BIOLOGICAL MOXOGRAPHS [56 Experiment' VII Rana clamitans Series 3557-3624 This experiment deals with the relative completeness of regenera- tion after successive removals within single individuals. Age and ex- ternal factors are not eliminated. A more complete description of the experiment is given under "Completeness of Regeneration." The tail length averaged approximately 17.0 mm. About one-half of the tail was removed at the first operation. At succeeding operations the cut came as near as possible to the border line between old and new tissue. The first removals came on October 23, 1911, the second on November 28, the third January 3, the fourth February 9, the fifth March 16 and the sixth April 4. At the time of the last removal the hind legs were just starting to grow. The data are given in Table 36. Tlie first regeneration interval is 36 days, the second 36, the third 37, the fourth 36 and the fifth 39 days. Each one of these is more than sufficient for the completion of the regenerative process. The individuals are divided into three sets, A, B, and C. A, with seven individuals, has no record for the first regen- eration ; the second regeneration is 9.8 mm., the third 9.3, the fourth 8.5 and the fifth 8.6. B, also with seven individuals, has no record for the first regeneration ; the second is 9.1 mm., the third 8.9, the fourth 7.2 and the fifth 7.8. C, with nineteen individuals, has a first regen- eration average of 8.6 mm., a second of 8.0, a third of 7.5, a fourth of 5.5 and a fiftli of 6.4. In all the cases there is a decrease in the amount regenerated with successive removal except for the fifth regeneration, which has in each case an increase over the fourth. It is probable that TABLE 36 Rana clamitans Series 3564-3624 Age factor not eliminated Completed successive regenerations compared First Second Third Fourth Fifth Catalog Number of regenera- regenera- regenera- regenera- regenera- Set number individ- tion tion tion tion tion uals 36 Days 36 Days 37 Days 36 Days 39 Days 3564 A to 3570 7 9.8 9.3 8.5 8.6 3578 B to 3584 7 9.1 8.9 7.2 7.8 3586 C to 3624 19 8.6 8.0 7.5 5.5 6.4 57] RATE OF REGESERATION—ZELEXY 57 the decrease is due to increase iu age. The increase from the fourth ta the fifth regeneration may be due to some special characteristic of the stage immediately preceding metamorphosis or it may merely indicate the existence of some uncontrolled external factor such as food or temperature. Discussion' The evidence shows clearly that wlieii the age factor is eliminated there is no decrease in rate of regeneration with successive removal. On the contrary the second regeneration is more rapid than the first up to the period of maximum rate. The second regeneration however passes its maximum sooner than does the first and after the tenth day the latter therefore catches up to the former in total amount regener- ated. There is no striking difi'erence between the second and the third regenerations but in each comparison the third has a slight advantage. When the successive regenerations in single individuals are com- pared, the rate decreases with successive removal. This decrease is undoubtedly due to the age factor. The possibility has suggested itself that the second regeneration starts out at a more rapid rate than the first because the cells at the cut surface were undergoing regenerative changes at the time of the new operation and can therefore start the process much faster than can the old cells at the first surface of regeneration. Following a first removal there is a considerable degree of reorganization of the cells at the cut surface, accompanied by active migration. During this period, which in Rana clamitans lasts two or three days, there is little or no mitotic cell division. Then follows a division period which reaches its maximum at seven to ten daj's. Its decline is associated with the oncoming of tissue differentiation (Sutherland 1915, iletealf 1915). A special study has been made of tlie ivlative rates of second regenerations from old cells following a cut inside of the first removal level and from new cells following a cut outside of the first level. Tliis comparison shows only a very slight difference in favor of tlie new cells and this is largely confined to the early stages, the period of cell migration. The i)eriod of increase in rate is the period of active cell iiiultipH- cation and the decline in rate is associated with cell differentiation. The second regeneration tlierefore reaches the period of differentiation slightly in advance of the first regeneration. Apart from the slowing due to age there is no indication of a limitation of the amount of new material that may be produced by regeneration. The actual limitation comes not from the using up of 58 ILLINOIS BIOLOGICAL MONOGRAPHS [58 regenerative or developmental energy or of determiners by repeated regeneration but from changes in the non-regenerating part associated with age. In another place there is a discussion of the possibility that there may be an eft'ect upon the rate of developmental processes in the organism as a whole due to continued regeneration of a part. This is studied particularly in connection with the effect of regeneration upon rate of metamorphosis in Amphibia. Regeneration studies in general and those on successive regener- ation in particular make it improbable that there is a definite number of cell generations between the fertilized egg and the end product, the differentiated cells. The possibilit}' that certain cells may remain in an early cell generation can not be wholl.y excluded as an explanation of at least a part of first regeneration phenomena. Under suitable stimu- lation such cells may be postvilated to take up development where it had left off. The definite descriptions of de-differentiations of cells as well as other facts of regeneration argue against this conclusion. The view that there can be no such definite immber of cell generations is strengthened by the facts of .successive regeneration. It does not seem probable that embryonic cells of an early cell generation can be held in reserve through repeated regenerations. The explanation of regeneration by the theory of duplicate sets of determiners meets difficulties in undimini.shed successive regenerations. The greater the number of repeated regenerations the greater the diffi- culties of explanation on this basis. Of course the difficulty does not hold for the hypothesis that everj' cell or nearly every cell contains a full set of determiners. The earlier appearance of the maximum rate in the second than in the first regeneration may be due to the more rapid progress of the cells in the early cell migration period alone or it may be due to the acceleration of the whole developmental cycle. Summary 1. The age factor was eliminated in Experiments I to IV. Ex- periments I and II deal with tadpoles of Rana clamitans and Experi- ments III and IV with larvae of Amblystoma punetatum. 2. In Experiment I approximately one-half of the tail was re- moved. At six days tlie average first regeneration length is 2.01 mm. and the average second regeneration length 2.18 mm. In five eases the first exceeds tlie second and in six tlie second exceeds the first. The corresponding specific lengths are 0.194 and 0.205. The first regen- eration exceeds the second in two sets, the second exceeds the first in eight and one is tied. The second regeneration has the advantage in all the comparisons. 59] RATE OF REGEXERATIOX—ZELENY 59 3. At eight days in Experiment I the average first regeueration length is 3.06 mm., and the second 3.42 mm. The fii'st exceeds the sec- ond in three sets and the second exceeds the first in seven. The corre- sponding average specific lengths are 0.298 and 0.323. In four sets the first regeneration exceeds the second and in six the second exceeds the first. The second regeneration has the advantage in all the com- parisons. 4. The advantage of the second regeneration over the first in Experiment I holds true of second regenerations from both old tissue and new tissue levels. 5. In Experiment II observations were made at the 1/10, 1/3, 1/2 and 2/3 levels in a sufficient number of individuals to yield valid data. Regeneration measurements were made at each of these levels 4, 6, 8, 10, 1214, 18 and 56 days after the operations. The second regen- eration at aU of them tends to be ahead of the first until the tenth day, after which the first regeneration catches up. The maximum rate for both regenerations is reached before this time and earlier for the second than for the first regeneration. 6. In Experiment III two-thirds of the tail was removed. A comparison of the first, second and third regenerations was made at 2, 4, 6, 8, 10 and 14 days.. At two days the first, .second and third regenerations average respectively 0.22, 0.25 and 0.26 mm. The cor- responding values at four days are 0.66, 0.75 and 1.00; at six days 1.36, 1.40 and 1.46 ; at eight days 2.18, 2.68 and 2.68 ; at ten days 3.55, 3.82 and 4.20; at fourteen days 5.34, 6.12 and 6.08. The advantage is in favor of the second and third regenerations as opposed to the first and of the third as opposed to the second. Individual comparisons at each of the different times as well as in the experiment as a whole show the same results. 7. The removed tails in the preliminary procedure of Experi- ment III furnish the data of Experiment IV and allow a comparison of the first and second regenerations at eleven days. The procedure is however subject to greater error than that of Experiments I to III. Twenty-five individuals for each regeneration give an average of 4.55 -0.11 mm. for the first regeneration and 4.50 -0.10 mm. for the second regeneration. The first regeneration is ahead of the second in ten cases, the second ahead of the first in twelve eases and three are equal. The two regenerations must be considered as essentially equal. 8. In Experiments V and VI the age factor is not eliminated. Successive regenerations in single individuals are compared. In Ex- periment V one-half of the tail in Arably stoma larvae was removed. In Experiment VI 60 per cent of the tail of toad tadpoles was removed. 60 ILLINOIS BIOLOGICAL MONOGRAPHS [60 The time intervals vary somewhat in each set but it is evident in botli cases that there is a decrease in rate of regeneration from the first to the third and fourth regenerations. This decrease is undoubtedly due to increase in age and not to successive removal. 9. In Experiment VII a comparison of the completeness of re- generation in single individuals of Rana clamitans shows a progressive decrease in amount regenerated from the first to the fourth regener- ation and an increase from the fourth to the fifth. In this experiment also the age factor is not eliminated and the decrease is probably due to increase in age. 61] RATE OF REGESERATIOS — ZELESY PART III THE EFFECT OF LEVEL OP THE CUT UPON THE RATE AND COMPLETENESS OF REGENERATION The present study gives a description of some experiments made to define more accurately than has been done the exact relation between the level of the cut and rate of regeneration and especially the relation of this factor to the other factors affecting rate and completeness of regeneration. The factor is one of great interest because if it is true that the ratio between length regenerated per unit time and length removed is a constant it follows that no matter how much material is removed regeneration is always completed in the same time. It is therefore of great interest to determine the extent to which this statement is true, to analyze the elements of the level factor and to determine its relation to other factors. ExPERiiiENT I Rax.v clamitans Series 3676-3765 The tadpoles were collected on December 9, 1911, and first removals were made in two-thirds of the individuals on December 22. A second removal was made in these individuals on January 8, and at the same time a first removal in the other one-third. Measurements were made four, six, eight, ten, twelve and a half, eighteen and fifty-six days after the operations of Januarj- 8. The first and second regenerations are treated separately and the second regenei'ations are taken up first because they have a larger number of individuals and therefore give the more uniform results. Second Regenerations The different amounts removed approximate 6, 10, IS, 31, 49 and 67 per cent of the tail length. There are four individuals at the lowest removal, averaging 1.5 mm., seven at the next, averaging 2.8 mm., five at the third with an average of 4.9 mm., ten at the fourth with 8.4 mm., eight at the fifth with 13.1 mm., and ten at the sixth M'ith 18.1 mm. The data are given in tables 37 to 40 and in graphic form in figures 4 to 17. The regenerated lengths at ten daj-s will be taken iip first because at this time the period of maximum rate has been passed and its full effect is represented. Differentiation of the tissues has begun but there 62 ILLINOIS BIOLOGICAL MONOGRAPHS [62 is still a considerable prodiiction of new cells by mitotic division except in the individuals with the two shortest removals in which the process is completed. The regenerated lengths for the six levels beginning witli the shortest removal are respectively 1.0, 1.3, 1.4, 2.3, 3.7 and 5.1 mm. The data are given in the last two columns of table 37. There is very dis- tinctly an increase in regenerated length with increase in removed length. Dividing the regenerated length by the removed length at each level, the fractions obtained are 1.0 1.3 1.4 2.3 3.7 5.1 T5' Ys 1^9' 'SA 131' isl' which give the specific regenerated lengths or lengths regenerated per unit of removed lengths. These values are 0.67, 0.46, 0.29, 0.28, 0.28 and 0.28. They show a remarkable constancy for removed lengths of 4.9 mm. and over. The relations between removed lengths and regenerated lengths are further shown in figure 4 which gives the removed lengths along the horizontal axis and the regenerated lengths parallel to the vertical axis. The plotted line of correlation between the two values is straight except for the two lowest removed lengths. The specific lengths are given in Figure 5 in which the removal lengths again are along the horizontal axis and the lengths regenerated per unit of removed length parallel to the vertical axis. The line of correlation is straight and parallel to the horizontal axis for the four highest removals. For these therefore the regenerated length is directly proportional to the removed length or in other words within these limits the same percent- age of the removed length is regenerated in each within the given time of ten days. The two lowest removed lengths give a higher specific rate than the others. They regenerate a higher percentage of the removed length within the given time. The ten day period is chosen as the fii'st example because it is the first one to receive the full benefit of the periods of maximum rate of regeneration, the periods during which rapid multiplication of cells takes place. The other periods give results which agree in general features after the first few days with those at ten daj's but depart from them in certain respects. The remaining periods will now be taken up in turn beginning with the shortest. During the first four days after the operation the rate of regenera- tion is slow, the new tissue being derived largely from migration of ceUs over the cut surface. Measurements of regeneration at this time are especially subject to error because of the small amount regenerated 63] RATE OF REGENERATION— ZELEKY 63 and because of irre^ilarity in the outer edge of the regenerating tissue. The regenerated lengths at four days are respectively 0.22, 0.39, 0.24, 0.42, 0.50 and 0.52 mm. These data are given in table 37 and are rep- mm. 5.0 1.5 2.8 4.9 8.4 13.1 — >- Lengths removed in mm. Figure 4 Rana clamitans Second regenerations Ten days 18.1 mm. 0.60 0.50 0.40 0.30 0.20 0.10 1.5 13.1 18.1 Figure 5 2.8 4.9 8.4 — >■ Lengths removed in mm. Raiia clamitans Second regenerations Specific lengths Ten days resented graphically in figure 6. Dividing the regenerated lengths by the removed lengths the fractions obtained are 0.22 0.39 0.24 0.42 0.50 0.52 islb 0.04 and 0.03. These There is on the wliole 1.50 2.80 4.90 8.40 13.10 giving specifie lengths of 0.15, 0.14, 0.05, 0.05, relations are represented graphically in figure 7. a slight increase in regenerated length with increase in removed length but this increase is not proportional to the amount removed so that the proportion regenerated decreases with increase in removed length. The 64 ILLINOIS BIOLOGICAL MONOGRAPHS [64 approach to equality in regeneration at this time is probably due to the fact that the new tissue is largely made up of migrating cells and there is not a striking difference in the extent of the migration at the different levels. The specific length of material regenerated after the smallest removals is greater than that regenerated after the larger removals not only at four days but also later. It is probable that the factors involved during the first few days of regeneration are quite different from those during later days. Following the injury there is a disintegration of injured cells associated with an active migration of the epidermal cells Figure 6 1.5 2.8 4.9 8.4 13.1 — >- Lengths removed in mm. Rana clamitans Second regenerations Four days mm. 0.20 0.10 o, Figure 7 1.5 2.8 4.9 8.4 — >- Lengths removed in mm. Ram clamitans Second regenerations Specific lengths Four days over the cut surface. There is practically no mitotic cell division. The rapid multiplication of cells comes later. These processes of cell migra- tion apparently are not essentially different at the different levels. They are local responses of the cells at the cut surface. With the appearance of rapid cell multiplication there is a marked difference at different levels though the shortest removals still show a greater specific length than the others probably because in their case the migrated cells make up a large percent of the total material of the new part. Between the end of the fourth and the end of the sixth day after the operation mitotic cell division becomes very rapid and the rate of regeneration for second regenerations reaches its maximum at a majority of the levels on the sixth day. At six days the regeneration for the six levels is respectively 0.62, 0.80, 0.70, 1.1, 1.7 and 2.3 mm., as shown in 65] RATE OF REGENERATION —ZELENY 65 Table 37. A graphic representation is given in Figure 8. There is a grad- ual increase with increase in removed length. The fractions obtained by dividing by the removed lengths are : "^0.62 0.80 0.70 1.1 1.7 2.3 1.5 2.8 4.9 8.4 13.1 18.1 They give specific lengths of 0.42, 0.30, 0.14, 0.13, 0.13 and 0.13. The smaller removals still have the larger specific lengths but with removals of 4.9 mm. and more there is an approach to constancy. The relations are shown graphically in Figure 9. 1.5 2.8 4.9 8.4 13.1 — >■ Lengths removed in mm. Figure 8 Rana clamitans Second regenerations Six days 0.40 0.30 0.20 0.10 Figure 9 1.5 2.8 4.9 8.4 — > Lengths removed in mm. Ram clamitans Second regenerations 13.1 Specific lengths Six days The rate of regeneration between the sixth and the eighth day for second regenerations is not quite as high as for the preceding period, but mitotic divisions are still very numerous and dift'erentiation of the cells is just beginning. At eight days the regenerated lengths are respectively 0.9, 1.1, 1.2, 1.8, 2.7 and 3.7 mm. as shown in table 37. The increase in regeneration with increase in removed length is evident. The relations are shown in Figure 10. Dividing by tlie removed lengths the fractions obtained are 0.9 1.1 1.2 1.8 2.7 3.7 1.5 2.8 4.9 8.4 13.1 18.1 66 ILLINOIS BIOLOGICAL MONOGRAPHS [66 giving the specific regenerations 0.60, 0.39, 0.24, 0.22, 0.21, 0.20. There is a graphic representation in Figiire 11. As before, the two shortest removals give the highest specific rates but beyond these there is an approach to constancy though there is still a slight decrease with increase in removal. The ten day values have already been given. Between ten and twelve and a half days after the operation there is j]o further growth in the case of the two shortest removals. In the two medium removals the process is completed at twelve and a half days. In the two longest removals there is still a small amount of mm. 4.0 Figure 10 1.5 2.8 4.9 8.4 13.1 — > Lengths removed in mm. Ram dantitans Second regenerations Eight days 18.1 mm. 0.60 0.50 0.40 0.30 0.20 0.10 Figure 11 13.1 18.1 1.5 2.8 4.9 8.4 — >- Lengths removed in mm. Rana clamitans Second regenerations Specific lengths Eight days proliferation after this time. At twelve and a half days the regenerated lengths are 1.0, 1.3, 1.6, 2.6, 4.4 and 6.2 mm. as shown in Table 38. The increase with increase in removed length is continuous. This is shown 67] - RATE OF REGEKERATION—ZELENY 67 in graphic form in figure 12. fractions obtained are 1.0 1.3 1.6 Dividing by tlie removed lengths the 2.6 4.4 6.2 1.5 'Yi 4.9 ~SA m '" 181 giving specific lengths of 0.67, 0.46, 0.33, 0.31, 0.34 and 0.34 graph for specific lengths is shown in figure 13. mm. 6.0 The There is still a fair Figure 12 •o ™™- £, 0.70 fe 0.60 I 0.50 <~ 0.40 0.30 0.20 0.10 ti 1.5 2.8 4.9 8.4 — >- Lengths removed in mm. Rana damitans Second regenerations Twelve and a half days Figure 13 1.5 2.8 4.9 8.4 — >- Lengths removed in mm. Rana damiians Second regenerations 13.1 18.1 Specific lengths Twelve and a half days apin-oach to constancy with removals of 4.9 mm. and above. The rela- tive increase in the case of the higher removals is due to the fact that regeneration is continuing in them after it has stopped in the others. 68 ILLINOIS BIOLOGICAL MOXOGRAPHS 168 Therefore the data after this time are values for the completeness of I'egeneration rather than for the rate. Between twelve and a half and eighteen days after tlie operation there is no further regeneration in the tails with the four shortest re- movals. Two of them even exhibit a decrease in size. The two longest removals show only a slight increase. At eighteen days the regenerated lengths are respectively 1.0, 1.2, 1.5, 2.6, 4.8 and 7.0 mm. as given in Table 38. The same data are represented in graphic form in Figure 14. Dividing by the removed lengths the fractions obtained are 1.0 1.2 1.5 2.6 4.8 7.0 1.5 2.8 4.9 8.4 13.1 18.1 giving specific lengths of 0.67, 0.43, 0.31, 0.31, 0.37 and 0.39. Tlie graph is shown in Figure 15. mm. 7.0 3.0 Figure 14 1.5 2.8 4.9 8.4 — > Lengths removed in mm. Ram clainitans Second regenerations Eighteen days At eighteen days there is very little regeneration at any of the levels and at some of them, especially the shorter removals, a consider- able absorption of regenerated material. Regeneration may therefore be considered as completed at this time. However the measurements for 56 days are given in order to show the changes. The regenerated lengths at that time are 0.9, 0.7, 1.6, 2.5, 5.2 and 7.1 mm. These data 69] mm. 0.70 0.60 0.50 0.40 0.30 0.20 0.10 Figure 15 RATE OF REGENERATION— ZELEXY 1.5 2.8 4.9 8.4 — >■ Lengths removed in mm. Ra)ia ciainitans Second regenerations Specific lengths Eight- een days are given in table 38 and are represented in grajiliie form in figiire 16. Dividing by the average removed lengths, which differ somewhat from the previous ones because of the death of certain individuals, the frac- tions obtained are 0.9 0.7 1.6 2.5 5.2 7.1 T5' 'ZG 1^ ~82 iZ2 '" 17^ mm. 7.0 1.5 2.6 4.9 S.2 — y Lengths removed in mm. Ram ciainitans Second regenerations Fifty-six days 70 ILLINOIS BIOLOGICAL MONOGRAPHS [70 giving specific regenerations of 0.60, 0.27, 0.33, 0.31, 0.39 and 0.40. Tliese are shown in the graph given in figure 17. Because of the absorption of regenerated tissue in the shorter removals and a slight growth in the longer ones the latter show a comparative increase in specific lengths. 0.60 0.50 0.40 0.30 0.20 0.10 Figure 17 1.5 2.6 4.9 8.2 — >- Lengths removed in mm. Ram clainitans Second regenerations Specific lengths Fifty- six days For a comparison of completeness of regeneration it is better to take the greatest lengths regenerated at each level rather than the amounts regenerated at any particular time because the shorter levels complete regeneration and begin to absorb the tissues sooner than do the longer ones. On this basis the greatest regenerated lengths at each of the six levels are, for the 1.5 mm. level 1.0 mm. reached at ten days, for the 2.8 level 1.3 mm. reached at ten days, for the 4.9 level 1.6 mm. reached at twelve and a half days, for the 8.4 level 2.6 mm. reached at twelve and a half days, for the 13.2 level 5.2 mm. reached at fifty-six days, and for the 17.9 level 7.1 mm. reached at fifty-six days. These data are given in tables 37, 38, 39 and 40 and in gr.iphic form in figure 18. At the last two levels there was a slight increase from eighteen to fift.y-six days but this almost certainly came during the early part of the period and the values are therefore completed values. Dividing by the removed lengths the fractions obtained are 1.0 1.3 1.6 2.6 5.2 and 7.1 1.5 2.8 4.9 8.4 13.2 17.9 giving specific lengths of 0.67, 0.46, 0.33, 0.31, 0.39 and 0.40. The graph is given in Figure 19. The high values for the two short levels are probably due to the fact that the cells migrating to the cut surface form RATE OF REGEXERATIOX — ZELENY mm. 7.0 6.0 J 2.0 Fig ure 18 •a mm. 0.70 c 0.60 be 0.50 « 0.40 to 0.30 0) 0.20 o 0.10 Figure 19 ness 1.5 2.8 4.9 8.4 13.2 — >- Lengths removed iu mm. Rana dainitans Second regenerations Completeness 17.9 1.5 2.8 13.2 4.9 8.4 • — > Lengths removed in mm. RiDUi dainitans Second regenerations Specific lengths 17.9 Complete- a large proportion of the total mass of the regenerated organ. Since apparentlj' the lengtli of this mass of cells is very much alike at all levels as indicated by the facts of the four day regenerations, the speeitic lengths for these short removals are greater than for the others. The high values of the two longest removals are due to a continuation of 72 ILLINOIS BIOLOGICAL MONOGRAPHS [72 regeneration at these levels after it has ceased at the others. At ten days the specific lengths regenerated are very nearly the same at all the levels except the first two. Rana clamitans TABLE 37 Series 3676-3765 Second regenerations Catalog number Removeo length 4 Days 6 Days 8 Days 10 Days Percent removed Average Regen- erated length Specific lengtli Regen- erated length Specific length Regen- erated length Specific length Regen- erated length Specific length 3676 1.3 0.27 0.60 0.9 1.0 3682 1.6 0.18 0.60 0.9 1.0 3730 1.6 0.39 0.75 0.9 0.9 6 3754 1.6 0.06 O.G.", 0.9 1.1 Average 1.5 0.22 0.15 0.62 0.42 0.9 0.60 1.0 0.67 3677 2.0 0.30 0.6 0.9 0.9 3696 2.1 0.48 0.8 1.0 1.1 3701 3.2 0.42 0.8 1.1 1.2 3713 2.8 0.36 0.8 0.9 0.9 10 3719 3.1 0.30 0.8 1.1 1.4 3749 2.8 0.48 0.6 1.2 1.4 3750 3.5 0.42 1.3 1.7 1.9 Average 2.8 0.39 0.14 0.80 0.30 1.1 0.39 1.3 0.46 3678 5.0 0.20 0.5 0.9 1.0 3684 5.5 0.15 0.7 1.2 1.4 3702 4.7 0.42 0.8 1.1 1.3 18 3720 4.6 0.06 0.3 1.3 1.8 3756 4.8 0.36 0.24 1.0 0.14 1.3 1.2 0.24 1.7 Average 4.9 0.05 0.70 1.4 0.29 ' ' 73] RATE OF REGEXERATION—ZELENY 73 Rana clamitans TABLE 37 (Continued) Series 3676-3765 Second regenerations Catalog number Removed lengtli 4 Days 6 Days 1 8 Days 10 Days Percent removed Average Regen- ciated length Specil'.c lengtli Regen- erated length .Specific length Regcn erated length .Specific length Regen- erated length Specific length 3679 8.4 0.30 0.7 1.4 1.9 3685 9.3 0.60 1.2 1.9 2.3 3697 7.3 0.48 1.2 1.7 2.2 3703 9.3 0.45 1.3 2.0 2.5 3715 7.9 0.24 0.9 1.7 2.3 31 3721 8.7 0.57 1.3 1.9 2.3 3733 8.5 0.36 1.0 1.9 2.4 3739 9.6 0,36 1.0 1.8 2.4 3751 6.7 0.45 1.1 1.7 2.1 3757 8.0 0.42 1.2 2.1 2.8 Average 8.4 0.42 0.05 1.1 0.13 1.8 . 0.22 2.3 0.28 3686 14.5 0.60 2.1 3.4 4.8 3698 14.9 0.50 1.5 3.3 4.3 3704 14.0 0.45 2.2 3.3 4.4 3716 12.7 0.39 1.7 2.4 3.4 3722 12.5 0.60 1.6 2.6 3.6 49 3740 13.9 0.30 1.1 2.1 3.0 3752 11.2 0.54 1.7 2.5 3.4 3758 11.0 0.60 1.5 2.2 2.9 Average 13.1 0.50 0.04 1.7 0.13 2.7 0.21 3.7 0.28 3680 16.0 0.60 1.9 3.0 4.2 3681 21.2 0.84 3.0 4.0 5.6 3687 19.7 0.54 3.6 5.6 6.0 3699 21.0 0.54 2.2 4.3 5.9 3705 17.6 0.72 2.0 3.6 4.8 3717 17.6 0.42 2.6 3.6 5.2 67 3723 18.4 0.30 2.3 3.7 5.3 3735 16.5 0.48 2.0 3.4 5.5 3741 16.0 0.30 1.9 3.0 5.4 3753 16.8 0.42 2.0 2.5 3.8 Average 18.1 0.52 0.03 2.3 0.13 3.7 0.20 5.1 0.28 74 ILLIXOIS BIOLOGICAL MONOGRAPHS [74 Rana clamitans TABLE 38 Series 3676-3765 Second regenerations Catalog number Kemoved length 1.3 12J Days 18 Days 56 Days Highest values Percent removed Average Regen- erated length Specific length Regen- erated length Specific length Regen- erated length Specific length Regen- erated length Specific length 3676 1.0 1.0 0.7 1.0 3682 1.6 1.0 1.0 1.1 1.1 3730 1.6 0.9 0.9 0.7 0.9 6 3754 1.6 1.2 1.2 1.1 1.2 Average 1.5 1.0 0.67 1.0 0.67 0.9 0.60 1.0 0.67 3677 2.0 0.9 0.9 0.7 0.9 3696 2.1 1.0 1.0 0.7 1.1 3701 3.2 0.9 0.9 0.5 1.2 3713 2.8 1.4 1.3 1.4 10 3719 3749 3750 3.1 2.8 3.5 1.4 2.0 1.3 1.3 2.0 1.3 0.9 1.4 2.0 1.9 Average 2.8 1.3 0.46 1.2 0.43 0.7 0.27 1.3 0.46 3678 5.0 1.1 1.1 1.1 3684 5.5 1.4 1.5 1.4 1.5 3702 4.7 1.3 1.3 1.3 1.3 18 3720 4.6 2.3 1.9 2.0 2.3 3756 4.8 1.8 1.6 1.6 1.8 1.6 Average 4.9 1.6 0.33 1.5 0.31 0.33 0.33 3679 8.4 1.9 2.1 2.1 3685 9.3 2.8 3.0 2.6 3.0 3697 7.3 2.4 2.3 2.1 2.4 3703 9.3 2.6 2.5 2.5 2.5 3715 7.9 2.6 2.6 2.8 2.8 3721 8.7 2.6 2.3 2.2 2.6 31 3733 3739 8.5 9.6 2.6 3.0 2.6 2.9 2.8 2.8 3.0 3751 6.7 2.4 2.5 2.3 2.5 3757 8.0 3.1 3.2 3.1 3.2 Average 8.4 2.6 0.31 2.6 0.31 2.5 0.31 2.6 0.31 75] RATE OF REGENERATION— ZELENY TABLE 38 (Continued) 75 Catalog number Removed length 12i Regen- erated length Days Specific length 18 Days 56 Days Highest values Percent removed Average Regen- erated length Specific length Regen. erated length Specific length Regen- erated length Specific length 3686 14.5 5.3 5.2 5.3 3698 14.9 5.0 5.4 5.4 5.4 3704 14.5 5.2 5.5 5.4 5.5 3716 12.7 4.2 5.1 4.4 5.1 3722 12.5 3.9 3.5 4.2 4.2 49 3740 3752 13.9 11.2 4.6 4.1 5.6 4.0 6.8 6.8 4.1 3758 11.0 3.6 4.1 4.9 4.9 Average 13.1 4.4 0.34 4.8 0.37 5.2 0.39 5.2 0.39 3680 16.0 5.2 6.4 6.6 6.6 3681 21.2 6.3 7.3 7.2 7.3 3687 19.7 6.6 7.0 7.0 3699 21.0 7.1 7.5 7.2 7.5 3705 17.6 6.4 6.2 6.0 6.4 67 3717 17.6 6.0 6.7 6.4 6.7 3723 18.4 6.5 8.1 8.3 8.3 3735 16.5 6.5 7.8 8.0 8.0 3741 16.0 5.8 6.9 7.0 7.0 3753 16.8 5.2 6.4 7.1 7.1 7.1 7.1 Average 18.1 6.2 0.34 7.0 0.39 0.40 0.40 TABLE 39 Rana clamitans Series 3676-3765 Summary Second regenerations Lengths regenerated at different levels at different times Percent of tail length removed Length removed In mm. Number of indi- viduals Days after operation 4 6 8 10 12/2 18 56 6 1.5 4 0.22 0.6 0.9 1.0 1.0 1.0 0.9 10 2.8 7 0.39 0.8 1.1 1.3 1.3 1.2 0.7 18 4.9 5 0.24 0.7 1.2 1.4 1.6 1.5 1.6 31 8.4 10 0.42 1.1 1.8 2.3 2.6 2.6 2.5 49 13.1 8 0.50 1.7 2.7 3.7 4.4 4.8 5.2 67 1«.1 10 0.52 2.3 3.7 5.1 6.2 7.0 7.1 76 ILLINOIS BIOLOGICAL MONOGRAPHS [76 TABLE 40 Rana clamitans Series 3676-3765 Summary Second regenerations Lengtlis regenerated at different levels at different times Percent of tail length removed Length removed in mm. Number of indi- viduals Days after operation 4 6 8 10 12/2 18 56 6 1.5 4 0.15 0.42 0.60 0.67 0.67 0.67 0.60 10 2.8 7 0.14 0.30 0.39 0.46 0.46 0.43 0.27 18 4.9 5 0.05 0.14 0.24 0.29 0.33 0.31 0.31 0.33 31 8.4 10 0.05 0.13 0.22 0.28 0.31 0.31 49 13.1 8 0.04 0.13 0.21 0.28 0.34 0.37 0.39 67 18.1 10 0.03 0.13 0.20 0.28 0.34 0.39 0.40 First Kegenerations The data for first regenerations are from a different set of indi- viduals than those for second regenerations. The two kinds of opera- tions were made on the same day. The general results obtained from the first regenerations are in full agreement with those obtained from the second regenerations but there is greater variability because of the smaller number of individuals. The average per cents of the tail length removed are respectively 6, 10, 17, 30, 48 and 62 for the six levels. The first of these has two individuals averaging 1.5 mm. of removed tail, the second five individuals with an average of 2.6 mm., the third three individuals with an average of 4.6, the fourth eight witli an average of 8.2, the fifth five with an average of 13.0 and the sixth five with an average of 16.7. The data for these experiments are given in Tables 41, 42, 43 and 44 and in Figures 20 to 35. The progress of a first regeneration is similar to that of a second except that the maximum is reached later in the case of first regenera- tions. In the present series the maximum specific rate for first regenera- tions comes between the sixth and the eighth day after the operation. A comparison of the two regenerations is made in the section on the effect of siaccessive removal. The change in rate during the process of regeneration is also discussed in a separate section. m RATE OF REGEXERATIOX—ZELEXY 77 The lengths regenerated during the first four days are respectively 0.27. 0.15, 0.51, 0.45, 0.46 and 0.51 for the six levels. There is no regular increase with removed length. The data are given in Table 41 and in Figure 20. The specific lengths regenerated are 0.17, 0.06, 0.11, 0.05, 0.03 and 0.03. They are shown in Table 41 and in Figure 21. As in tlie case of the second regeneration the shortest removals have the largest proportional amounts. This first period being the period of cell 1.5 13.0 1^ Figure 20 ;.6 4.6 8.2 — >- Lengths removed in mm. Rana claiiiilans First regenerations Four days 16.7 migration with very little cell division it is probable that the length of the material furnished in this way, as measured along the main axis of the individual, is independent of the level of the cut. The area of the cut surface of course is greater at the more proximal than at the more distal levels so that the actual total mass of regenerated material is greater at the deeper levels. At six days the rate of first regenerations is rapidly increasing, the maximiim rate coming between six and eight days. The lengths regenerated at six days are respectively 0.47, 0.5, 1.0, 1.1, 1.6 and 1.6 mm. mm. 0.20 0.10 1.5 13.0 Figure 21 2.6 4.6 8.2 — >- Lengths removed in mm. RaiM damitans First regenerations Specific lengths 16.7 Four days 78 ILLINOIS BIOLOGICAL MONOGRAPHS [78 They are shown in Table 41 and in Figiire 22. There is in general an increase with increase in removed lengtli but the first is not proportional to the second. The specific lengths are 0.30, 0.18, 0.22, 0.13, 0.12 and 0.09 as shown in Table 41 and Figure 23. The shorter removals still have proportionately the greater regenerations. ■g mm. rt 2.0 1.5 2.6 4.6 S.2 i; — > Lengths removed in mm. Figure 22 Rana claiiiitans. First regenerations Si.x days 0.30 0.20 0.10 •5 Figure 23 1.5 2.6 4,6 8.2 — >- Lengths removed in mm. Rana damitans First regenerations Specific lengths Six days The maximum rate of regeneration is reached between the sixth and the eighth day. The regenerated lengths at eight days are 0.8, 0.7, 1.5, 1.7, 2.6 and 3.0 mm. The data are shown in Table 41 and Figure 24. With one exception there is increase in regenerated length with increase in removed length. The specific regenerated lengths are 0.53, 0.27, 0.33, 0.21, 0.20 and 0.18 as shown in Table 41 and Figure 25. The shortest removals have the greatest specific regenerations but at 8.2 mm. and above there is an approach to constancy. 79] RATE OF REGENERATION— ZELENY 79 Figure 24 1.5 13.0 2.6 4.6 8.2 — >- Lengths removed in mm. Rana daiiiitans First regenerations Eight days M 0.60 0.50 0.40 0.30 0.20 0.10 1.5 13.0 2.6 4.6 8.2 — >- Lengths removed in mm. Figure 26 Ra)i-i claiiiilans First regenerations Specific lengths Eight days Between the eighth and the tenth day there is a rapid decrease in rate associated with tissue differentiation. The regenerated lengths at ten days are 0.9, 1.0, 1.7, 2.3, 3.8 and 4.5 mm. as show^l in Tal)l(> 41 and Figure 26. There is an uninterrupted increase in regeneration with increase in removed length. The specific lengtlis are 0.58, G.3b. 0.37, 0.28, 0.29 and 0.27 as shown in Table 41 and Figure 27. 80 ILLINOIS BIOLOGICAL MONOGRAPHS [80 Between teu and twelve and a half days regeneration is slow, reach- ing its end for the two shortest removals at the latter day. The regen- erated lengths at twelve and a half days are 0.9, 1.2, 1.8, 2.6, 4.7 and mm. 5.0 1.5 2.6 4.6 8.2 13.0 — >■ Lengths removed in mm. Figure 26 Rana damitans. First regenerations Ten days mm. 0.60 0.50 0.40 0.30 0.20 0.10 1.5 2.6 4.6 8.2 — >- Lengtlis removed in mm. Rana damitans First regenerations Specific lengtlis Ten days 5.8 mm. as shown in Table 42 and Figure 28. There is a steady increase with increase in removed length. The specific lengths are 0.61, 0.46, 0.39, 0.31, 0.36 and 0.35 as shown in Table 42 and Figure 29. There is an approach to constancy in the four largest removals. 811 RATE OF REGENERATION —ZELENY 81 Between twelve and a half and eighteen days the regenerated material is decreasing in the case of the two shortest removals, has made no progress in the third and in tlie three longest removals the increase is very sliglit. The data therefore are of value more particularly in con- nection with the problem of the relative completeness of regeneration from the different levels. The regenerated lengths at eighteen days are mm. 6.0 13.0 16.7 1.5 2.6 4.6 8.2 — >- Lengths removed in mm. Figure 28 Rana damitans First regenerations Twelve and a lialf days „ mm 01 0,70 C3 0.60 13 0.50 u 0.40 ji 0.30 hfi a 0.20 13.0 16.7 1.5 2.6 4.6 8.2 — >■ Lengths removed in mm. Figure 29 Rana claiiiitaits First regenerations Specific lengths Twelve and a half days 82 ILLINOIS BIOLOGICAL MONOGRAPHS 182 0.9, 1.1, 1.8, 2.7, 5.5 and 6.8 mm. as shown in Table 42 and Figure 30. There is a regular increase with increase iu removed length. The specific lengths are 0.60, 0.42, 0.39, 0.33, 0.42 and 0.40 as shown in Table 42 and Figure 31. Though there are irregularities the specific lengths approach constancy at all levels except the most distal one. Between eighteen and fifty-six days there is practically no increase mm. 7.0 4.0 3.0 1.0 1.5 2.6 4.6 8.2 — >■ Lengths removed in mm. Figure 30 ' Rana dainiians. First regenerations Eighteen days mm. 0.70 0.60 0..50 0.40 0.30 0.20 0.10 Figure 31. Ran days 1..5 2.6 4.6 8.2 13.0 — y Lengths removed in mm. ' daiiiitans First regenerations Specific lengths 83] RATE OF REGENERATION— ZELENY 83 in regenerated length and some absorption of material especially ^vith the shorter removals. The regenerated lengths at fifty-six days are 0.7, 1.1, 1.5, 2.5, 5.5 and 6.9 mm. as shown in Table 42 and Figure 32. There is a regular increase in regeneration from the shortest to the longest removal. The specific lengths are 0.45, 0.42, 0.34, 0.30, 0.42 and 0.41 as shown in Table 42 and Figure 33. These data are of value only for mm. 7.0 Figure 32 •a mm. c 0.50 • Lengths removed In mm. Ram clamitans First regenerations Fifty-six days 13.0 16.7 1.5 2.6 4.6 8.2 — > Lengths removed in mm. ure 33 Rami damitans First regenerations Specific lengths Fifty-six days 84 ILLINOIS BIOLOGICAL MONOGRAPHS [84 a comparison of completeness of regeneration but for such a comparison it is better in some ways to compare the regenerations at the time when absorption lias not begun. The greatest average regenerated length attained for the 1.5 mm. level is 0.9 mm. at ten days, for the 2.6 level 1.2 at twelve and a half days, for the 4.6 level 1.8 at twelve and a half days, for the 8.2 level 2.7 at eighteen days, for the 13.0 level 5.5 at eighteen days and for the 16.7 level 6.9 at fifty-six days. There is an ■ uninterrupted increase from the shortest to the longest removal in com- plete amount regenerated. This is shown graphically in Figure 34. The 7.0 Figure 34 -a mm. 0.70 - Lengths removed in mm. Rana dainitans First regenerations Completeness Figure 35 5 2.6 4.6 8.2 — >- Lengths removed in mm. Rana clamitans First regenerations Completeness Specific lengths 85] RATE OF REGENERATION— ZELENY 85 completed regenerations are less than the removed lengths. The specific completed regenerated lengths obtained as before by dividing by the removed lengths are 0.61, 0.46, 0.39, 0.33, 0.42 and 0.41, as shown in table 42 and figure 35. The greater specific lengths from the shortest removals are probablj' due as in the case of the second regenerations to the fact that a greater proportion of their substance is made up of cells that have migrated over the cut surface during the first stages of regeneration. This migrated material is not essentially different in axial length at the diiferent levels. The largest removals have a greater spe- cific length than the medium ones because regeneration continues at the former levels after it has ceased at the latter. On the whole there is no essential difference between the results obtained from first regenerations and those obtained from second regen- erations. The latter give the more regular data because the averages are taken from a larger number of individuals. Rana clamitans TABLE 41 Series 3676-3765 First regenerations Catalog number Removed length 4 I Regen- erated length )ays Specific length 6 Days 8 Days 10 Regen- erated length Days Percent removed Average Regen- erated length Specific length Regen- erated length Specific length Specific length 3706 1.4 0.24 0.54 0.9 1.0 6 3742 1.7 0.30 0.40 0.7 0.8 Average 1.5 0.27 0.17 0.47 0.30 0.8 0.53 0.9 0.58 3688 2.5 0.12 0.3 0.3 0.7 3707 3.2 0.24 0.8 1.1 1.4 3724 2.6 0.06 0.5 0.8 1.1 10 3743 2.5 0.03 0.1 0.4 0.8 3760 3.1 0.30 0.6 0.9 1.1 Average 2.6 0.15 0.06 0.5 0.18 0.7 0.27 1.0 0.38 86 ILLINOIS BIOLOGICAL MONOGRAPHS [86 87] RATE OF REGEXERATIOX —ZELENY 87 Rana clamitans TABLE 42 Series 3676-3765 First regenerations Catalog number Re- moved length in mm. 12* Days 18 Days 56 Days Percent removed Average Regen- erated length mm. Specific length Regen- erated length Specific length Regen- erated length mm. Specific length 3706 1.4 1.0 0.9 0.7 6 3742 1.7 0.9 0.9 0.7 Average 1.5 0.9 0.61 0.9 0.60 0.7 0.45 3688 2.5 0.9 0.9 0.7 3707 3.2 1.4 1.3 0.7 3724 2.6 1.4 1.4 1.2 10 3743 2.5 1.0 1.0 1.7 3760 3.1 1.2 1.1 1.1 Average 2.6 1.2 0.46 1.1 0.42 1.1 0.42 3708 5.3 2.3 2.3 1.8 3726 4.3 1.4 1.4 1.4 17 3762 4.1 1.7 1.8 1.4 Average 4.6 1.8 0.39 1.8 0.39 1.5 0.34 3690 9.7 2.6 2.7 2.2 3709 8.8 3.2 3.4 3.3 3727 8.3 2.2 2.2 2.2 3745 10.0 4.4 4.8 4.2 3744 6.0 1.8 1.7 30 3761 3763 6.6 8.5 2.2 2.9 2.3 3.1 1.8 3689 6.3 1.6 1.7 1.4 Average 8.2 2.6 0.31 2.7 0.33 2.5 0.30 3710 12.3 3.9 3.9 3.9 3728 12.8 4.8 5.4 5.8 3746 13.3 5.7 7.0 6.8 48 3764 14.6 5.3 6.8 6.5 3765 12.2 3.9 4.5 4.5 Average 13.0 4.7 0.36 5.5 0.42 5.5 0.42 3692 16.8 4.3 5.0 5.2 3693 17.2 6.5 7.3 6.6 3711 17.0 7.0 7.7 8.3 62 3729 16.1 5.5 6.7 6.4 3749 16.2 5.6 7.1 7.8 6.9 Average 16.7 5.8 0.35 6.8 0.40 0.41 ILLJXOIS BIOLOGICAL MOXOCRAPHS TABLE 43 Rana clamitans Series 3676-3765 Sumniaiy First regenerations Lengths regenerated at different levels at different times Percent of tail length removed Length removed in mm. Number of indi- viduals Days after operation 4 6 8 10 12/2 18 56 6 1.5 2 0.27 0.15 0.5 0.5 1.0 0.8 0.9 0.9 0.9 0.7 10 2.6 5 0.7 1.0 1.2 1.1 1.1 17 4.6 3 0.51 1.5 1.7 1.8 1.8 1.5 30 8.2 8 0.45 1.1 1.7 2.6 2.3 2.6 2.7 2.5 48 13.0 5 0.46 1.6 3.8 4.7 5.5 5.5 62 16.7 5 0.51 1.6 3.0 4.5 5.8 6.8 6.9 TABLE 44 Rana clamitans Series 3676-3765 Summary First regenerations Specific lengths regenerated at different levels at different times Percent of tail length removed Length removed in mm. Number of indi- viduals Days after operation 4 6 8 10 12/2 18 56 6 1.5 2 0.17 0.30 0.53 0.58 0.61 0.60 0.45 10 2.6 5 0.06 0.18 0.27 0.38 0.46 0.42 0.42 17 4.6 3 0.11 0.22 0.33 0.37 0.39 0.39 0.34 30 8.2 8 0.05 0.13 0.21 0.28 0.31 0.33 0.30 48 13.0 5 0.03 0.03 0.12 0.20 0.29 0.36 0.42 0.40 0.42 62 16.7 5 0.09 0.18 0.27 0.35 0.41 Experiment II Amblystoma punctatum Series 4600-5052 The eggs were hatched on March 29 to April 4, 1913. Opera- tions on the tail were made on May 7 in numbers 4600-4752 and on May 10 in numbers 4800-5052. The removed lengths were approximately Vio. Vr,> Vs! V2 and V, of the tail length. Measurements of the regen- erated tissue were made at 2, 4, 6, 8-9, 10-11, 13, 15-16 and 17-18 days after the operation. The data are given in Tables 45 to 54 and in Figures 36 to 51. The salamander larvae are much more irregular in their regeneration as well as in ordinary growth than frog tadpoles. The measurements 89) RATE OF RECEXERATION—ZELENY 89 in the present experiment were made on killed individuals so that only a single regeneration measurement is made in a single individual. This procedure also tends toward a greater variability in the data. Tlie number of individuals in any particular measurement also is less tlian for the second regeneration of frog tadpoles. Xotwithstanding all these unwelcome factors the general features of regeneration are similar to those for the tadpole experiment. The regenerated lengtli at any time is approximately i^roportional to the removed length. It is true even in the earliest measurements. As for frog tadpoles the shorter removals have proportionately a larger regeneration than the others at practically each time of measurement. The approach to equality in specific lengths is true only of tlie lengths of removal equal to one-fifth or more of the tail length. At two days the regenerated lengths are respectively 0.10, 0.15, 0.15, 0.47 and 0.53 mm. for the five levels of removal. They give specific lengths of 0.07, 0.07, 0.04, 0.08 and 0.06 as shown Tabh> 45 aiul Figures 36 and 37. At four days tlie regenerated lengths are 0.12, 0.15, 0.30, 0.41 and 0.40 mm. and the specific lengths 0.11, 0.07, 0.07, 0.07 and 0.05 as shown in Table 46 and Figures 38 and 39. At six days the regenerated lengths are 0.32, 0.47, 0.62, 0.70 and 1.02 mm. and the specific lengths 0.30, 0.20, 0.16, 0.12 and 0.11 as shown in Table 47 and Figures 40 and 41. At eight to nine days tlie regenerated lengths are 0.40, 0.65, 0.80, 1.40 and 1.52 mm. and the specific lengths 0.44, 0.28. 0.23, 0.23 and 0.19 as shown in Table 48 and Figures 42 and 43. At ten to eleven days the regenerated lengths are 0.50, 0.63, 1.54, 2.22 and 2.22 mm. and "the specific lengths 0.62, 0.26, 0.43, 0.41 and 0.27 as shown in Table 49 and Figures 44 and 45. At thirteen days the regenerated lengths are 0.78, 0.92,- 1.74, 2.40 and 3.60 mm and the specific lengths 0.74, 0.43. 0.48, 0.44 and 0.48 as shown in Table 50 and Figures 46 and 47. At fifteen to sixteen days the regenerated lengths are 0.80, 1.30, 1.37, 2.80 and 3.80 mm. and the specific lengths 0.67, 0.61, 0.40, 0.48 and 0.54 as shown in Table 51 and Figures 48 and 49. At seventeen to eighteen days the regenerated lengths are 0.70, 1.40, 1.60, 3.80 and 4.67 mm. and the specific lengths 0.67, 0.62, 0.41, 0.66 and 0.57 as shown in Table 52 and Figures 50 and 51. A summary of regenerated lengths is given in Table 53 and of specific regenerated lengths in Table 54. Since the experiment was closed at eighteen days and since the measurements at different times were made on different individuals it is not possible to make as accurate a comparison of completeness of 90 ILLINOIS BIOLOGICAL MONOGRAPHS [90 regeneration as in the case of the frog tadpoles. For the three shortest removals regeneration is probably completed at this time biit this is not true for the two longest ones. In this respect as in others there is an agreement with the former experiment. The percent of the removed tail that is regenerated is greater for all levels than in the frog tadpoles. It is probable also that if the longest removals had been allowed to com- plete their regenerations their specific regenerations as in the case of the frog tadpoles would have been shown to be greater than those for medium levels. TABLE 45 Amblystoma punctatum. Series 4600-.50.52. Average tail length =10.9 mm. Regeneration: 2 days Percent of tail length Catalog Removed Regenerated Specific removed number length length length Average mm. mm. regenerated 14 5022 1.5 0.1 Average . 1.5 0.10 0.07 4641 2.3 0.2 4741 1.9 0.1 20 4841 2.2 0.1 5050b 2.3 0.2 Average 2.2 0.15 0.07 4811 3.9 0.3 4911 3.3 0.1 32 5012 3.3 0.05 Average 3.5 6.0 0.15 0.04 4601 0.3 4801 6.0 0.7 53 5001 5.4 0.4 Average 5.8 0.47 0.08 4631 9.5 0.7 4831 9.1 0.6 81 5032 7.9 0.3 Average 8.8 0.53 0.06 91] RATE OF REGENERATION— ZELENY 91 The data from botli experiments show that except for very short removals the length regenerated in a given time is approximately pro- portional to the length removed. TABLE 46 Amblystoma punctatum. Series 4600-.^0.'2. Average tail length=10.9 mm. Regeneration: 4 days Percent of tail length Catalog Removed Regenerated Specific removed number length length length Average mm. mm. regenerated 4622 0.9 0.1 4722 0.8 0.2 4822 1.5 0.1 10 4922 o.s 0.1 5024 1.6 0.1 Average 1.1 0.12 0.11 4742 2.4 0.1 21 4842 2.3 0.2 Average 2.3 0.15 0.07 4612 3.7 0.3 4712 3.6 0.2 37 4812 4.5 0.4 5012 4.2 0.3 Average 4.0 0.30 0.07 4602 6.0 0.2 4702 6.0 0.05 4802 6.2 0.7 55 4902 6.0 0.5 5004 5.9 0.6 Average 6.0 9.3 0.41 0.07 4632 0.4 4732 8.5 0.2 4832 10.9 0.6 76 4932 5.9 0.4 5034 7.1 0.4 Average 8.3 0.40 0.05 92 ILLINOIS BIOLOGICAL MONOGRAPHS [92 Amblystoma punctatum. TABLE 47 Series 4600-5052. Average tail length=10.9 mm. Regeneration: 6 days Percent of tail lengtli Catalog Removed Regenerated Specific . removed number length length length Average mm. mm. regenerated 4623 1.0 0.3 4723 0.8 0.2 10 4923 1.6 0.6 1.1 0.2 0.32 Average 1.1 0.30 4643 2.1 0.7 4743 2.2 0.2 21 4843 2.1 0.4 4943 2.8 0.6 Average 2.3 0.47 0.20 4613 3.6 0.6 4713 2.9 0.6 4820b 4.4 0.8 35 4913 4.5 0.6 5013 3.6 0.5 Average 3.8 0.62 0.16 4603 6.1 0.4 4703 6.7 0.4 54 4803 5.3 1.4 5003 5.6 0,6 Average 5.9 0.70 0.12 4633 8.8 0.6 4733 8.2 0.9 82 4833 10.6 1.8 5033 7.9 0.8 Average 8.9 1.02 0.11 93] RATE OF REGENERATIOS — ZELESY 93 Amblystoma punctatum Regeneration TABLE 48 Series 4600-5052 Average tail length=10.9 8-9 days (8 for 4800-5052, 9 for 4600-4752) Percent of tail length removed Average Catalog number Removed length mm. Regenerated length mm. Specific length regenerated 8 4724 4824 502G 0.7 1.1 1.0 0.2 0.7 0.3 Average 0.9 0.40 0.44 21 4644 4844 4944 5045 2.3 2.3 2.2 2.3 0.6 1.0 0.7 0.3 Average 2.3 0.65 0.28 31 4614 4624 4714 4814 4914 5016 3.0 3.0 3.4 3.7 3.2 4.3 0.8 0.4 1.2 1.0 0.6 0.8 Average 3.4 0.80 0.23 56 4604 4704 4804 4904 5006 6.0 6.3 5.9 5.4 6.8 1.8 1.7 1.4 0.9 1.2 Average 6.1 1.40 0.23 75 4634 4734 4834 4934 8.3 8.0 9.0 7.5 2.1 1.3 1.6 1.1 Average 8.2 1.52 0.19 94 ILLINOIS BIOLOGICAL MONOGRAPHS [94 TABLE 49 Amblj'stoma punctatum Series 4600-5052 Average tail lengtli=10.9 mm. Regeneration: 10-11 days (10 for 4800-5052, 11 for 4600-4752) Percent of tail length Catalog Removed Regenerated Specific removed number length length length Average mm. mm. regenerated 4725 0.6 0.3 4825 0.8 0.6 7 5027 1.0 0.6 Average 0.8 0.50 0.62 4746 2.6 0.4 4845 2.6 1.0 23 5046 2.2 0.5 Average 2.5 3.0 0.63 0.26 4620b 1.9 4715 3.0 1.8 4815 4.3 2.0 33 4920 3.6 1.2 5017 3.9 0.8 Average 3.6 1.54 0.43 4605 5.2 2.8 4705 4.6 1.4 4805 6.3 2.8 50 4910b 5.0 1.9 5007 6.0 2,2 Average 5.4.. 2.22 0.41 4735 7.5 2.5 4835 9.4 2.9 74 4935 7.6 1.3 5037 8.1 2.2 Average 8.1 2.22 0.27 95] RATE OF REGENERATION — ZELENY 95 Amblystoma punctatum. Series 4600-5052. Average tail length= Regeneration: 13 days Percent of tail length removed Average Catalog number Removed length mm. Regenerated length mm. Specific length regenerated 10 4626 4726 4830b 4926 5028 1.4 1.0 1.0 1.0 0.9 0.8 0.6 0.7 0.9 0.9 Average 1.1 0.78 0.74 19 4646 4745 4846 4946 1.9 2.2 2.6 1.9 1.4 0.8 1.0 0.5 Average 2.1 0.92 0.43 32 4616 4716 4816 4916 5018 3.6 3.2 3.9 3.7 3.7 1.8 2.0 2.5 1.6 0.8 Average 3.6 1.74 0.48 50 4706 4806 4910 5008 6.4 5.8 5.0 4.5 2.3 ■ 2.9 2.5 1.9 Average 5.4 2.40 0.44 69 4636 4740b 4936 5038 8.1 7.7 7.4 6.7 3.4 4.7 3.5 2.8 Average 7.5 3.60 0.48 96 ILLINOIS BIOLOGICAL MONOGRAPHS [96 TABLE 51 Amblystoma punctatum. Series 4600-5052. Average tail length^lO.9 mm. Regeneration: 15-16 days (15 for 4800-5052, 16 for 4600-47g2) Percent of tail length removed Average Catalog number Removed length mm. 1.2 1.2 Regenerated length mm. Specific length regenerated 11 4927 5029 0.8 0.8 Average 1.2 0.80 0.67 19 4647 4747 4847 4950b 5049 1.9 2.7 2.4 1.9 1.8 1.5 1.7 1.0 1.2 1.1 Average 2.1 1.30 0.61 32 4617 4717 4917 5019 3.1 3.5 3.2 4.1 1.4 1.3 1.8 1.0 Average 3.5 1.37 0.40 53 4607 4807 4817 5009 5.7 5.7 5.2 6.6 2.7 2.9 2.6 3.0 Average 5.8 2.80 0.48 64 4937 7.0 3.8 Average 7.0 3.80 0.54 97] RATE OF REGENERATION —ZELENY 97 TABLE 52 Amblystoma punctatura. Series 4600-5052. Average tail Iength=10.9 mm. Regeneration: 17-18 days (18 for 4800-5052, 17 for 4600-4752) Percent of tail lengtti Catalog Removed Regenerated Specific removed number length length length Average mm. mm. regenerated 4828 1.0 0.8 10 4929 1.1 0.6 Average 1.0 0.70 0.67 4648 2.1 1.5 4749 2.6 2.1 4848 2.1 1.1 20 4949 2.2 1.3 5050 2.2 1.0 Average 2.2 1.40 0.62 4718 3.9 1.6 36 Average 3.9 1.60 0.41 4608 5.4 4.2 4708 6.6 4.1 53 4808 5.4 3.1 Average 5.8 3.80 0.66 4838 9.4 5.0 4939 6.4 4.5 76 5040 9.0 4.5 Average 8.3 4.67 0.57 ILLINOIS BIOLOGICAL MONOGRAPHS TABLE 53 Amblystoma punctatum Series 4600-5052 Summary Regenerated lengths (Tables 45 to 52) Percent of tail length removed Average Length removed mm. Average Average le ngth regenerated In mm. 2 Days 4 Days 0.12 0.15 0.30 6 Days 8-9 Days 10-11 Days 13 Days 0.78 0.92 15-16 Days 17-18 Days 10 1.1 0.10 0.32 0.40 0.65 0.80 1.40 1.52 0.50 0.63 0.80 0.70 21 2.2 0.15 0.47 1.30 1.40 34 3.7 0.15 0.62 1.54 1.74 1.37 1.60 53 5.8 0.47 0.41 0.40 0.70 2.22 2.40 3.60 2.80 3.80 74 8.1 0.53 0.94 2.22 3.80 4.70 TABLE 54 Amblystoma punctatum Series 4600-5052 Summary Specific lengths regenerated (Tables 45 to 52) Percent of Length removed mm. Average Ave rage sp ecific regenerated lengths tail length removed Average 2 Days 4 Days 6 Days 8-9 Days 10-11 Days 13 Days 15-16 Days 17-18 Days 10 1.1 0.07 0.11 0.30 0.20 0.16 0.43 0.62 0.74 0.67 0.67 21 2.2 0.07 0.04 0.06 0.28 0.26 0.43 0.61 0.62 34 3.7 0.07 0.23 0.43 0.48 0.44 0.48 0.40 0.41 53 5.8 0.08 0.07 0.12 0.11 0.23 0,41 0.27 0.48 0.54 0.66 74 8.1 0.06 0.15 0.19 0.57 99] RATE OF REGENERATION —ZELEKY 99 2.2 3.5 5.8 Lengths removed in mm. Figure 36 Amblystoiiia i^iiiictafuiii Lengths regenerated Two days mm. 0.10 - Lengths removed in mm. Amblystoma fmctatuiu Specific lengths regenerated Two days 0.67 c 1.1 2.3 4.0 6.0 8 kJ — > Lengths removed in mm. Figure 38 Amblystoiiia fuiutatuiii Lengths regenerated Four days S Figure 39 1.1 2.3 4.0 6.0 8.3 — >■ Lengths removed in mm. Amblystoma puiictatuin Specific lengths regenerated Four days 100 ILLINOIS BIOLOGICAL MONOGRAPHS HOC >4 Figure 40 o £ mm. oi to 0.30 Si 0.20 » 0.10 ® Figure 41 mm. 1.33 0.67 1.1 2.3 3.8 .0.9 — >- Lengtlis removed in mm. Amblystoma puuctatiiiii Lengtlis regenerated Six dayy 1.1 2.3 3.8 5.9 8.9 — >- Lengths removed in mm. Amblystoma punctatuin Specific lengths regenerated Six days 0.9 6.1 8.2 2.3 3.4 — >- Lengths removed in mm. Figure 42 Aiiiblystoiiia puiirtatitin Lengths regenerated Eight to nine days 'S mm. g 0.50 g 0.40 £ 0.30 5 0.20 S 0.10 I ^ 0.9 2.3 3.4 6.1 8.2 — > Lengths removed in mm. Figure 43 Amblystoma pnnctatuin Specific lengths regenerated Eight to nine days 101] RATE OF REGEXERATION—ZELENY 101 1.33 0.8 8.1 Figure 44 2.5 3.6 5.4 — >■ Lengths removed in mm. Amblystoiua l^unctatum Lengths regenerated Ten to eleven days en mm. 0.60 0.50 0.40 0.30 0.20 0.10 0.8 2.5 3.6 5.4 — > Lengths removed in mm. Figure 45 Ainhlysioma punctatum Specific lengths regenerated eleven days 102 ILLINOIS BIOLOGICAL MONOGRAPHS [102 mm. 3.33 2 2.00 1.1 2.1 3.6 5.4 7.5 — >- Lengths removed in mm. Figure 46 Ainblystoiiia pniiitatiini Lengths regenerated Thirteen days mm. 0.70 0.60 0.50 0.40 0.30 0.20 0.10 Figure 47 1.1 2.1 3.6 5.4 7.5 — y Lengths removed in mm. Amblystoma punctatum Specific lengths regenerated Thirteen days 103] RATE OF REGEXERATIOX—ZELEKY 103 mm. 4.00 1.2 2.1 3.5 5.8 7.0 — >- Lengths removed In mm. Figure 48 Amblysioma pnnctatuiK. Lengths regenerated Fifteen to sixteen S 09"0 £ 0.50 0.30 1.2 7.0 2.1 3.5 5.8 — >■ Lengths removed in mm. Figure 49 Amblystoma i>unctat\im Specific lengths regenerated Fifteen to sixteen days 104 ILLINOIS BIOLOGICAL MOXOGRAPHS [104 mm. 4.67 2.67 1 1.0 2.2 3.9 5.8 8.3 — >- Lengths removed in mm. Figure 50 Ainblystoma punctatuin Lengths regenerated Seventeen to eight- een days W mm. 0.70 0.60 0.50 0.40 0.30 0.20 0.10 1.0 5.8 2.2 3.9 — > Lengths removed in mm. Figure 51 Ainblystoma punctatuiti Specific lengths regenerated to eighteen days I 105] RATE OF REGEXERATION—ZELENY 105 Discussion That the level of the cut has an important influence upon the rate of regeneration has been made out by a number of investigators (Spal- lanzani 1768, King 1898, Morgan 1906, Stockard 1908, Ellis 1909, Morgu- lis 1909a, b, and others). Their work indicates that regenerations from deeper levels are on the whole more rapid that from more superficial ones. The data obtained from the present experiments confirm tliis conclusion and make possible a further analysis of the relation. They show that in the regeneration of the tail of amphibian larvae there is a striking relation between the level of the cut and the rate of regeneration. Within wide limits the length regenerated is directly proportional to the distance of the cut surface from tlie original tip of the tail. Within these limits therefore regeneration at any particular time after the operation has the same degree of completeness from all levels of injury. An anal.ysis of the progress of the regeneration brings out the fact that two distinct periods are to be recognized in rate of regeneration in its relation to level of the cut. During the first two to four days after the operation regeneration is confined to cell migration from the old tissues without cell di\'ision. During this period in the frog tad- poles there is no essential difference in lengtli regenerated at the differ- ent levels and the specific rate is therefore much greater after shorter than after longer removals. In the second period with the initiation of rapid cell multiplication the rate of regeneration is greater the deeper the level and furthermore is directly proportional to the length removed. As soon as the bulk of material produced by cell division is considerably greater than that which was prodviced by cell migration there is an approach to constancy in specific length regenerated. This holds for all except the shortest removals. After the shortest removals the total regeneration is so small in amount that a large part of it is made up of the original migrated material. Therefore from these levels the spe- cific regenerated lengtlis are greater than from the deeper levels even at a late period of regeneration. A further complication is introduced by the fact that regeneration is not complete. Only a certain per cent of the removed length is re- placed and the end of the process is reached .sooner after the shorter than after the longer removals. From the deepest levels regeneration is still proceeding when it has stopped from the medium and shallowest ones. When the process is completed in all cases the specific length is therefore slightly greater after botli the longest and the shortest re- movals tlian after medium ones. As to the cause of the tliffei-ence in rate at the different levels 106 ILLINOIS BIOLOGICAL MONOGRAPHS [106 little more can be said than that it does not seem to be due to inherent differences in the cells at the different levels. If differentiation in the tail proceeded from the tip toward the base, the more rapid rate fi-om the more basal levels might be explained by the more embryonic char- acter of the cells at tliese levels. As the tip is approached the material would become more and more inert. There is however no evidence tliat differentiation proceeds in this way in this case. The progressive increase in rate with depth of level of tlie cut is undoubtedly due to reactions which involve a more central control, a co-ordination of the functional activity as a whole. The period of cell migration probably is only slightly subject to such control. It is a period in which the response is largely local in character and there is correspondingly little if any difference at the different levels. The rate of cell division which is the important factor during the period of rapid increase in length is however undoubtedy under central control. Summary 1. In frog and salamander larvae with removed tail lengths of one-fifth to two-thirds, the general rule holds that the length regenerated in a given time is proportional to the length removed, or in other words the length regenerated per unit of removed lengtli is a constant. 2. An analysis of the data shows however that this applies only to the material produced by active cell division. 3. During the first four days, in frog tadpoles, when the regener- ating part is made up almost entirely of cells that have migrated from the old tissues without division there is no such relation between length i-emoved and length regenerated. The length of new material at this time is not strikingly different for the different levels and the process seems to be a local response of the cells to the injury. The length regenerated per unit of removed length is greater at this time for the shorter than for the longer removals. 4. Since comparatively a large part of the regenerating material after the shorter removals is made up of migrated cells even at the later periods it follows that the specific regenerations from these levels are greater than from the deeper ones. 5. During the later periods the specific regenerated lengths tend to be higher after both the shortest and the longest removals than after medium ones. In tlie case of the shortest ones this is due to the rela- tivel.v large part of the whole regenerated tail that is made up of mi- grated cells. In the case of the longest removals it is due to the fact that regeneration continues for a time after it has stopped in the medium ones. 107] RATE OF REGENERATION— ZELENY 107 6. It does not seem probable that the differences in length regener- ated at different levels can be due to differences in tlie original character of the cells involved in the process. Such a well graduated difference in cell capacities is difficult to conceive. The process must be iinder a more central control, probably connected with general functional activity. 108 ILLIXOIS BIOLOGICAL MOXOGRAPHS [108 PART IV THE CHANGE IN RATE OF REGENERATION DURING THE REGENERATIVE PROCESS The present experiments were undertaken in extension of previous studies on the change in rate throughout the regenerative cycle. This previous work showed that the increase in amount of material during regeneration follows the general rule of increase during an ordinary life cycle. The rate is at first very slow, then increases very rapidly to a maximum, then declines rapidly at first and then more and more slowly as zero is approached. Frog tadpoles and salamander larvae were used in the present study. Large tadpoles of Bona clamitans which remained fairly con- stant in size during the course of the experiments were found to be the most satisfactory. The results obtained from them were uniform enough for an analysis of the change in rate. The salamander larvae showed a great variation in rate from day to day apparently associated with external factors such as food and temperature. The data obtained from them are however of interest in comparison with the frog tadpole results. The experiments will be taken up in turn beginning with the series containing the largest number of individuals and giving the most iini- form results. Experiment I Rana clamitans Second regenerations of the TAIL Series 3676-3765 The tadpoles were collected on December 9, 1911 and first remov- als were made on December 22 and second removals on January 8. Measurements were taken 4, 6, 8, 10, 12i/o and 56 days after the opera- tion. The operations were made at six different levels, the removals approximating 6, 10, 18, 31, 49 and 67 per cent of the tail length. The first of these removals averaged 1.5 mm. and four individuals with completed measurements are available, the next averaged 2.8 mm. with seven individuals, the third 4.9 mm. with five, the fourth 8.4 mm. with ten, the fifth 13.1 mm. with eight and the sixth 18.1 mm. with ten individuals. The rates per day for each level during each period are given in table 55 and in graphic form in figure 52. The maximum 109] RATE OF REGEXERATIOX—ZELEXY 109 2 5 7 9 11>4 1">U — >■ Days after the operation Figure 52 Rates of second regenerations of the tail per day at different times after the operation for six different levels Rana damitans The removed lengths are 1.5, 2.8, 4.9. 8.4, 13.1 and 18.1 mm. 110 ILLINOIS BIOLOGICAL MONOGRAPHS rate is reached during the period between four and six daj's at three of the levels and between six and eight daj's at the other three. The rise in rate is very rapid and the decline also rapid. As discussed in the preceding section on the effect of the level of the cut, the rate of regeneration increases with depth of the level and the increase is such that in general the specific length or length regener- ated per unit of removed lengtli is approximately a constant. A reduc- tion of the rates to specific rates therefore gives an opportunity for averaging the different levels together. The resultant average is based upon a sufficiently large number of individuals to give a considerable degree of smoothness in the curve of rate. The data for specific rate 2 nd 2 5 7 9 11% 151/4 — > Days after the operation Figure 53. Specific rates of first and second regenerations at different times after tlie operation Rana claniitans Tail regeneration Upper figure, second regenerations; lower, first regenerations. are given in Table 56. The average specific rates for all six levels to- gether are 0.019 mm. during the to 4 day period, 0.066 during the 4 to 6 day period, 0.0.51 for 6 to 8 days, 0.03.3 for 8 to 10 days, 0.017 for 10 to 12i/o days, 0.001 for 12yo to 18 days and —0.001 for 18 to 56 days. This change in rate is represented graphically in the upper part of Figure 53. For the four deepest levels the averages are given in a separate column of Table 56. Tliey exclude the two lowest levels which Ill] RATE OF REGEXERATIOX—ZELEA'V depart considerably from the others iu specific rate. There is how- ever no essential difference in the two sets of values as regards the form of the rate curve. The change in rate of regeneration or acceleration of rate from any period to the succeeding one is shown iu Table 57 in which the period of change is represented by the middle days of the two periods which are being compared. The average of all the levels shows the acceler- ation to be +0.095 mm. from the 2 to tlie 5 day period, — 0.015 for 5 to 7 days, —0.030 for 7 to 9 days, —0.058 for 9 "to 111,4 days, —0.028 for 111/4 to 151/4 days and — 0.001 for I514 to 37 days. It is only between the first two periods that acceleration of rate is a plus quantity. Dur- ing all the others it is minus, the most rapid rate of decrease coming between 9 and 111/4 days. The accelerations of specific rate are more reliable measures for obtaining averages including tlie different periods. Such values are given in Table 58 and iu graphic form in Figure 54. They give a result in the relation of the periods to each other essentially similar to that above. The average accelerations of specific rate are -(-0.014 for the Figure 54. tail in 6 8 10 1314 26 — >- Days after the operation Acceleration of specific rate First and second regenerations of the Rana claiiiitans Unbroken line=First regeneration Broken Iine= Second regeneration. 2 to 5 day periods, — 0.004 for 5 to 7 days, — 0.009 for 7 to 9 days, —0.0085 for 9 to II14 days, —0.003 for II14 to I514 days and 0.000 for 1514 to 37 days. The first period is the only one with a plus accel- eration. The greatest minus acceleration comes between the 7 and the 9 day periods instead of 9 to 111/4 days. Averaging only the regener- ations for the four deepest levels which show a constant specific rate 112 ILLINOIS BIOLOGICAL MONOGRAPHS [112 the values are respectively +0.011, 0.000, —0.005, —0.006, —0.004 aud 0.000, putting the greatest rate of decrease between the 9 and the III/4 day periods. An examination of the curves of specific rate and a comparison with the facts of histogenesis shows that acceleration of rate is a plus quantitj'' only during the period before active differentiation of the cells, i. e. until the end of the fifth or seventh day. As soon as tissue differentiation is fairly begun the retarding influence is apparent and by the nintli to eleventh days when muscle fibres and other cells are in full process of differentiation the negative acceleration is at its height. Following the percentage increment method used by Minot (1908) for ordinar.y growth and using length instead of weight becaiise the latter could not be determined with sufficient accuracy the results given in Table 59 are obtained. The values for the six periods excluding the first one are 106, 28, 12, 5 and 0. The regenerated material present at the end of four days is made up almost wholly of cells that have migrated from the old tissues and have not as yet undergone division. After the fourth day the additions to regenerated material are almost wholly the result of cell division. From the end of the fourth to the end of the sixth day the material is on the average more than doubled in length each day. After this time the percentage increment decreases rapidly. The change from period to period is represented in graphic form in Figure 55. The curve is a logarithmic one quite similar to that obtained 5 7 9 11% 15% — >- Days after the operation Figure 55 Percentage increment per day at different periods after the opera- tion First and second regenerations of the tail of Rana danntans U^^ broken line=first regeneration. Brolten line=second regeneration. 113J RATE OF REGENERATION — ZELENY 113 hy ilinot for growth. It should however be poiuted out that both regeueration aud ordiuary growth undoubtedh' have a verj' rapidly aseeudiug branch of the curve if the very beginuings of the processes are included. TABLE 55 Rana clamitans Series 3676-3765 Second regenerations Rate of regeneration of tail per day at different times during the regenerative process for six different levels Percent of tail length removed 6 10 18 31 49 67 Length removed in mm. 1.5 2.8 4.9 8.4 13.1 18.1 No. of in- dividuals 4 7 5 10 8 10 Days 0- 4 0.05 0.10 0.06 0.10 0.12 0.13 4- 6 0.20* 0.20* 0.23 0.34 0.40 0.91* 6- 8 0.14 0.15 0.25* 0.35* 0.50* 0.70 8-10 0.05 0.10 0.10 0.25 0.12 0.50* 0.70 10-12 V4 0.00 0.00 0.08 0.28 0.44 12%-18 0.00 —0.02 —0.02 0.00 0.07 0.01 0.15 18-56 0.00 —0.01 0.00 0.00 00 114 ILLIXOIS BIOLOGICAL MONOGRAPHS Rana clamitans Series 3676-3765 Second regenerations Specific rates at different levels at different times Percent of tail length removed 6 10 18 31 49 67 18.1 Average of all levels Average of four longest remov- Length removed in mm. 1.5 2.8 ' 4.9 8.4 13.1 No. of in- dividuals 4 7 5 10 8 10 Days 0- 4 0.037 0.035 0.012 0.012 0.010 0.045* 0.007 0.019 0.010 4- 6 0.135* 0.080* 0.045 0.040 0.050* 0.066* 0.045* 8-10 0.035 0.035 0.025 0.030 0.035 0.035 0.051 0.042 6- 8 0.090 0.045 0.050* 0.045* 0.040 0.040 0.033 0.032 10-121/2 0.000 0.000 0.025 0.015 0.030 0.030 0.017 0.025 12%-18 0.000 —0.005 —0.004 0.000 0.005 0.001 0.009 0.001 —0.001 0.002 18-56 —0.002 —0.004 0.001 0.000 0.000 0.000 115] RATE OF RECEXERATIOK—ZELEKY 115 TABLE 57 Rana clamitans Series 3676-3765 Second regenerations Acceleration ot rate of regeneration of tail per day at different times during the regenerative process for six different levels Percent of tail length removed Length removed in mm. No. of in- dividuals Middle ot periods Days 2- 5 9-11% ll%-15y4 151/4-37 -0.05' -0.03 -0.04* -0.02 0.00 -0.00 10 18 2.8 4.9 7 5 -I-0.03* + 0.06* —0.02 -1-0.01 —0.02 -0.07* —0.04* —0.01 —0.00 —0.02 —0.00 + 0.00 + 0.08* +0.00 —0.05 —0.06* —0.03 —0.00 + 0.09' +0.05 0.00 —0.10* —0.05 —0.00 +0.26' -0.07 -0.01 Average of all levels 116 ILLINOIS BIOLOGICAL MONOGRAPHS [116 TABLE 58 Rana clamitans Series 3676-3765 First regenerations Acceleration of specific rate of regeneration of the tail Percent of tail lengtli removed Length removed In mm. No. of in- dividuals Days 2- 5 9-1114 —I ii%-i5y4 15%-37 6 10 18 31 49 67, 18.1 Average of . all levels 1.5 2.8 4.9 8.4 13.1 4 7 5 10 8 10 + 0.033* + 0.011* +0.012* +0.010* +0.007* + 0.014* +0.014* —0.020 —0.007 +0.002 0.000 + 0.004 —0.006 —0.004 —0.027* —0.007 —0.014* —0.006 0.000 0.000 —0.009* —0.013 -0.014* —0.002 —0.007* -0.008* —0.007 —0.008 0.000 0.000 —0.004 -0.004 —0.004 —0.004 —0.003 0.000 0.000 0.000 0.000 0.000 —0.001 0.000 Average of four deepest levels +0.011' 117] RATE OF REGENERATION — ZELENY 117 TABLE 59 Rana clamitans Series 3676-3765 First regenerations Percentage increment of regenerating tail per day during each time period for six different levels Percent of tail length 6 10 18 31 49 67 removed Average Length of removed 1.5 2.8 4.9 8.4 13.1 18.1 ail in mm. levels No. of in- dividuals 4 7 5 10 8 10 Days 4- 6 91 53 96 142 80 175 106 6- 8 23 19 36 32 29 30 28 8-10 5 9 8 14 19 19 12 10-12% 6 5 8 9 5 12>4-18 —2 —1 2 2 18-56 —0 —1 +0 —0 + ( +0 118 ILLISOIS BIOLOGICAL MONOGRAPHS [118 Experiment II Raxa clamitans First regenerations op the tail, Series 3676-3765 The tadpoles were collected on December 9, 1911, and the tail remov- als were made on January 8. Measurements were taken 4, 6, 8, 10, 12%, 18 and 56 days after the opei'ations. The operations were at six levels approximating 6, 10, 17, 30, 48 and 62 per cent of the original tail length. For the first of these levels only two individuals with an average removal of 1.5 mm. are available, for the second five individuals with 2.6 mm., for the third three with 4.6 mm., for the fourth eight with 8.2 mm., for the fifth five with 13.0 mm. and for the sixth five with 16.7 mm. The rates of regeneration per day are given in table 60 and the graphs for the rates in Figure 56. Tlie specific rates are given in Table 61. Averaging these values so as to include all the different levels for each period the specific rates are 0.018 for to 4 days, 0.046 for 4 to 6 days, 0.057 for 6 to 8 days, 0.037 for 8 to 10 days, 0.026 for 10 to I21/2 days, 0.002 for 121/0 to 18 days and — 0.001 for 18 to 56 days. The graph is shown in the unbroken line in Figure 53. Using only the four deepest levels the average specific rates are respectively 0.013," 0.042, 0.045, 0.036, 0.025, 0.006 and 0.000 giving essentially the same form of curve as for the average of all levels. The accelerations of rate are shown in Table 62 and the accelerations of specific rate in Table 63 and in the unbroken line of Figure 54. The average accelerations of rate per day are respectively -|-0.07S, 0.000, — 0.022, —0.042, —0.025 and 0.000 mm. The average accelerations of specific rate including all levels are respectively +0.011, — 0.001, —0.007, —0.0075, —0.003 and 0.000 and including only the four deepest levels, +0.009, 0.000, —0.004, —0.005, —0.003 and 0.000. As for sec- ond regenerations the only plus acceleration is between 2 and 5 days and the most rapid deci-ease takes place between 9 and 11^/4 days. The percentage increments per day are shown in Table 64 and in the unbroken line of Figure 55. The values are respectively 98, 29, 17, 6, 1 and percent per day giving approximately the same form of curve as for second regenerations. In general the first regenerations agree with the second but on the whole the second regenerations reach their maximum earlier and are more rapid than the first up to the time of maximum rate. The first are more rapid than the second after the maximum. 119] RATE OF RECESERATIOX—ZELESY 119 2 7 9 llli 13 ■* — >■ Days after the operation Figure 56 Rates of first regenerations of the tail per day at different times after the operation for six different levels Ranci damitans T^'ie removed lengths are 1.5, 2.6, 4.6, 8.2, 13.0 and 16.7 mm. 120 ILLINOIS BIOLOGICAL MONOGRAPHS [120 TABLE 60 Rana clamitans Series 3676-3765 First regenerations Rate of regeneration of tail per day at different times during the regenerative process for six different levels Percent of tail length removed 6 10 17 30 48 62 Length removed in mm. 1.5 2.6 4.6 8.2 13.0 16.7 No. of individuals 2 5 3 8 5 5 Days 0- 4 0.07 0.02 0.12 0.10 0.12 0.12 4- 6 0.10 0.20* 0.25* 0.35* 0.55* 0.55 6- 8 0.15* 0.10 0.25* 0.30 0.50 0.70 S-10 0.05 0.15 0.10 0.30 0.40 0.75* 1 10-121/2 0.00 0.00 0.08 0.04 0.12 0.36 0.52 121/2-18 —0.02 0.00 0.02 0.15 0.18 18-56 -0.01 0.00 —0.01 —0.01 0.00 0.00 121] RATE OF REGEXERATION — ZELENY 121 TABLE 61. Rana clamitans Series 3676-3765 Specific rates at different levels Second regenerations at different times Percent of tail length removed 10 17 30 48 62 16.7 5 Average of all levels Average of four longest remov- Length removed in mm. 1.5 2.6 4.6 8.2 13.0 No. of in- dividuals 2 5 3 8 5 Days 0- 4 0.042 0.015 0.027 0.012 0.007 0.007 0.018 0.013 4- 6 0.065 0.040 0.055* 0.040* 0.045* 0.030 0.046 0.057* 0.042 6- 8 0.115* 0.045 0.055* 0.040* 0.040 0.045* 0.045* 8-10 0.025 0.055* 0.020 0.035 0.045* 0.045* 0.037 0.036 10-121/2 0.015 0.040 —0.007 0.010 0.015 0.035 0.040 0.026 0.002 —0.001 0.025 121/2-18 —0.002 0.000 0.004 0.011 0.009 0.006 18-56 —0.004 —0.001 —0.001 —0.001 0.000 0.000 0.000 ILLIXOIS BIOLOGICAL MOXOCR.-IPHS [122 TABLE 62 Rana clamitans Series 3676-3765 First regenerations Acceleration of rate of regeneration of tail per day at different times during the regenerative process for six different levels Percent of tail length removed 6 10 17 30 48 62 Average of all levels Length re- moved in mm. 1.5 2.6 4.6 8.2 8 13.0 16.7 No. of in- dividuals 2 3 5 5 Middle of periods Days. 2- 5 + 0.01 + 0.06* + 0.04* + 0.08* + 0.14* +0.14* +0.078* 5- 7 +0.02* —0.05* 0.00 —0.07* —0.02 —0.02 + 0.07 0.000 7- 9 —0.05* + 0.02 0.00 —0.05* + 0.02 —0.022 9-11% —0.02 —0.03 —0.02 —0.07* —0.02 —0.09* -0.06 —0.042* Il'4-15i4 0.00 -0.02 —0.01 —0.02 —0.04 -0.025. 1514-37 —0.00 0.00 —0.00 —0.00 0.00 + 0.00 —0.000 H 123] RATE OF REGEXERATIOS—ZELESY 123 TABLE 63 Rana clamitans Series 3676-3765 First regenerations Acceleration of specific rate of regeneration of the tail Percent of tail length removed 6 10 18 31 49 67 Average of . all levels Average of four deepest levels Length removed in mm. 1.5 2.8 4.9 8.4 13.1 18.1 No. of in- dividuals 4 7 5 10 S 10 Days 2- 5 f0.007 -1-0.023* f0.009* -fO.OlO* -fO.Oll* -1-0.008* -1-0.011* 4-0.009* 5- 7 fO.013* —0.019* 0.000 —0.002 -0.002 -1-0.004 —0.001 0.000 7- 9 —0.033* 4-0.008 —0.015* 0.000 —0.004* -1-0.001 -0.007 —0.004 9-11% —0.013 —0.012 —0.004 —0.002 -0.009* —0.002 —0.005* —0.007* —0.005* ll%-15y4 0.000 —0.008 —0.002 —0.003 -0.004 —0.003 —0.003 15»4-37 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000 124 ILLINOIS BIOLOGICAL MONOGRAPHS [124 TABLE 64 Rana clamitans Series 3676-3765 First regenerations Percentage increment of regenerating tail per day during each time period for six different levels Percent of tail length removed 6 10 17 30 48 62 Average of all Length re- moved in 1.5 2.6 4.6 8.2 13.0 16.7 mm. levels No. of in- dividuals 2 5 3 8 3 5 Days 4- 6 33 200 50 87 110 110 " 98 6- 8 30 20 25 27 31 44 29 8-10 6 21 7 18 23 25 17 10-121/2 8 2 5 9 12 6 121/2-18 -2 1 3 3 1 18-56 . —0 —0 ■ —0 +0 Experiment III Rana clamitans . First and second regenerations OP THE tail Series 3628-3675 For comparison with the data of experiments I and II it is of interest to note the results obtained from this entirely different series of the same species which was designed primarily for the comparison of first and second regenerations. A full description of tlie experiment is given in the section on the effect of successive removal upon the rate of regeneration. The data of specific value for present purposes are given in Table 65. Measurements were made only at six and at eight days after the operation. Fifty percent in length of the tail was removed in both first and second regenerations. Twenty- one individuals are available for first and sixteen for second regenerations. The rates per day are 0.52 mm. for 6 to 8 days for first regener- ations as compared with 0.50 for the same period in Experiment II and 0.62 for second regenerations as compared with 0.50 in Experiment I. The specific rate per day for 6 to 8 days for first regenerations is 0.049 125] RATE OF REGEXERATIOX—ZELEXV 125 and for second regenerations 0.057 as compared with 0.050 for forty eight percent removals in the first regenerations of Experiment II and 0.050 for forty nine percent removals in the second regenerations of Experiment I. The percentage increments per day are 26 for first regenerations as compared with 29 in Experiment II and 28 for second regenerations as compared with 28 in Experiment I. The close agreement of these values taken from a comparatively large number of individuals strengthens the conclusion as to the validity of the comparisons at different periods and levels in experiments I and II. TABLE 65 Rana clamitans Series 3628-3765 First and second regenerations of the tail Six and eight days No. of individ- uals Total length Tail length mm. Percent of length re- moved Length moved Regen- erated length Six days Regen- erated length Eight days Rate per day Spe- cific rate Percent- age in- crement per day First regeneration 21 16 32.7 21.4 50 10.6 10.9 2.01 3.06 0.52 0.049 26 Second regeneration 33.4 21.8 50 2.18 3.42 0.64 0.057 28 Experiment IV Amblystoma punctatum Tail Series 4600-5052 Operations were made at five levels approximating 10, 21, 34, 53 and 74 per cent of the original tail length. The removed lengths average respectively 1.1, 2.2, 3.7, 5.8 and S.l mm. Measurements were made 2, 4, 6, 8-9, 10-11, 13, 14-15 and 16-17 days after the operation. The rates per day for each of the levels at each of the different times are given in Table 66. The specific rates are shown in Table 67. The averages for all the levels at each of the time periods are respectively 0.032 mm. for to 2 days, 0.004 for 2 to 4 days, 0.053 for 4 to 6 days, 0.039 for 6 to 8 days, 0.064 for 8.4 to 10.3 days, 0.043 for 10.3 to 13.0 days, 0.012 for 13.0 to 15.2 days and 0.019 for 15.2 to 17.3 days. As in the case of other salamander experiments the data are more irregular than those for the frog tadpoles because of the susceptibility of the salamander larvae to factors which have not so far been brought under control. The character of the food is probably an important factor. The greatest rate comes between 8.4 and 10.3 days after the operation for three of the five levels and also for the average of all levels. This is later than 126 ILLIXOIS BIOLOGICAL MONOGRAPHS [126 the maxiimini for the frog tadpole which comes between four and six days for second regenerations and between six and eight days for first regenerations. The period of decline in rate is also more extended in these salamander larvae than in the frog tadpoles of Expei'imeuts I and II. On account of the irregularity of the data it is not possible to study the acceleration of rate for the present data. The percentage increments per day are given in Table 68. The values for the seven time periods are respectively 8, 71, 20, 23, 14, 4 and 7. The greatest percentage increment comes between 4 and 6 days as in the case of the frog tadpoles. An earlier period, that be- tween two and four days is represented here. During this period the percentage increment is low. If this value can be accepted the curve here includes the very steep ascending portion discussed above. The irregularities in rate to which the salamander larvae are subject and the fact that the low value during this period does not ajspear in all the salamander experiments however makes the interpretation doubtful. TABLE 66 Amblystoma punctatum Series 4600-5052 Rate of regeneration of tail per day at different times during the regenerative process for five different levels Percent of tail removed 10 21 34 53 74 Length removed, in mm. 1.1 2.2 3.7 5.8 8.1 Days 0-2 0.05 0.07 0.07 0.23 0.26 2-4 0.01 0.00 0.07 —0.03 —0.06 4-6 0.10* 0.16 0.16 0.14 0.27 6-8.4 0.03 0.07 0.07 0.29 0.21 8.4-10.3 0.05 —0.01 0.39* 0.43 0.37 10.3-13.0 0.10* 0.11 0.07 0.07 0.51* 13.0-15.2 0.01 0.17* —0.17 0.18 0.09 15.2-17.3 —0.05 0.05 0.11 0.48* 0.41 127] RATE OF REGENERATION— ZELENY 127 TABLE 67 Amblystoma punctatum Series 4600-5052 Specific rates of regeneration of the tail at different levels at different times after the operation Percent of tail removed 10 21 34 53 5.8 74 Average of all levels Length re- moved in mm. 1.1 2.2 3.7 8.1 Days 0-2 0.035 0.035 0.020 0.040 0.030 0.032 2-4 0.020 —0.005 0.015 —0.005 —0.005 0.004 4-6 0.095 0.070 0.045 0.025 0.030 0.053 6-8.4 0.054 0.033 0.029 0.046 0.033 0.039 8.4-10.3 0.100* —0.011 0.105* 0.095* 0.042 0.064* 10.3-13.0 0.044 0.063 0.019 0.011 0.078* 0.043 13.0-13.2 —0.032 0.081* —0.036 0.018 0.0G9 0.027 0.012 15.2-17.3 0.000 0.005 005 0.014 0.019 128 ILLINOIS BIOLOGICAL MONOGRAPHS [128 TABLE 68 Amblystoma punctatum Series 4600-5052 Percentage increment of regenerating tail per day during each time period for five different levels Percent of tail removed 10 21 34 3.7 53 74 Average of all .levels Length re- moved in mm. 1.1 2.2 5.8 8.1 Days 2-4 10 50 —6 —12 8 4-6 83* 10 107* 53* 35 77* 71* 6-8.4 16 12 42* 20 20 S. 4-10.3 13 — 2 49 31 24 23 10.3-13.0 27 17 5 3 23 14 13.0-15.2 1 —6 19 —10 8 3 4 15.2-17.3 4 8 17 11 7 Experiment V Amblystoma punctatum Tail Series 4101-4540 The experiment consists of the regenerations of removed halves of the tail without additional injury in some individuals and with an additional removal of tlie two forelegs in others. Measurements were made at nine periods, 2, 4, 6, 8, 10, 12, 14, 16 and 19 days after the operation. The rates of regeneration are given in Table 69. The number of individuals for most of the levels is five. The full data are discussed in the section on the effect of degree of injury. The average rate for each of the different times shows that the maximum comes during the eight to ten day period. The high value for the greater degree of injury at 14 to 16 days is due to the death during that period of the two individuals with the lowest values. The result agrees very well with the maximum rate in Experiment IV. The percentage increments are given in Table 70. The highest value comes during the two to four day period followed by decrease with but little irregularity. 129] RATE OF REGENERATION— ZELENY 129 Experiment VI Amblystoma punctatum Tail Series 3962-4004 First, second and third regenerations after removal of approximately one-half of the tail were studied. The complete data are given in the section on the effect of successive removals. Measurements were made at 2, 4, 6, 8, 10 and 14 days. The rates per day are given in Table 71. The maximum rate comes between 8 and 10 days agreeing with the other data for regeneration of the tail in salamander larvae. The percentage increments are given in Table 72. The highest rate comes at tlie earliest period, between two and four days, and is followed by a rapid and then a slower decrease. TABLE 69 Amblystoma punctatum Series 4101-4540 Rate of regeneration per day of tail at different times during the regenerative process for two degrees of injury Period of regeneration Days Middle of period Days after operation Rate of regeneration per day for each period Average rate One-half tail One-half tail + fore-legs 0-2 1 0.17 0.13 0.15 2-4 3 0.19 0.27 0.23 4-6 5 0.29 0.25 0.27 6-8 7 0.37 0.47 0.42 8-10 9 0.69* 0.50 0.59* 10-12 11 0.46 0.46 0.46 12-14 13 0.23 0.37 0.30 14-16 15 0.37 0.53* 0.45 16-19 17% 0.16 0.03 0.09 130 ILLIXOIS BIOLOGICAL MONOGRAPHS [130 TABLE 70 Amblystoma punctatum Series 4101-4540 Percentage increment per day of regenerating tail at different times during the regenerative process for two degrees of injury Days 2 to 4 4 to 6 6 to 8 8 to 10 10 to 12 12 to 14 14 to 16 Percentage increment per day during each period One-half tail One-half tail-|- fore-legs Average TABLE 71 Amblystoma punctatum Series 4101-4540 Rate of regeneration per day at different times during the regenerative process Period of regeneration Days Middle of period Days after operation Rate of regeneration per day during each period Average First Second Third 0-2 1 0.11 0.12 0.13 0.12 2-4 3 0.22 0.25 0.37 0.28 4-6 5 0.35 0.32 0.18 0.28 6-8 7 0.41 0.64* 0.66 0.57 8-10 9 0.68* 0.57 0.76* 0.67* 10-14 12 0.45 0.57 0.47 0.50 131] RATE OF REGEXERATIOX—ZELEXY 131 TABLE 72 Amblystoma punctatum Series 3962-4004 Percentage increment per day at different times during the regenerative process Days Percentage increment per day during each period Average First Second Third 2 to 4 100 100 142 114* 4 to 6 53 43 18 38 6 to 8 30 45 48 41 8 to 10 31 21 28 27 10 to 14 12 15 11 13 Experiment VII Amblystoma punctatum Series 4101-4540 Forelegs The experiment consists of the study of the rate of regeneration of single completelj' removed fore-legs under three degrees of injury to the individual: without additional injury, with the other fore-leg re- moved at the same time and with the other fore-leg plus one-half of the tail removed . Measurements were made at 2, 4, 6, 8, 10, 12, 14, 16 and 19 days. The rates of regeneration are given in Table 73. The maxi- mum rate does not come until the 14 to 16 period. The percentage increments are given in Table 74. The highest value comes during the 2 to 4 day period. There is a gradual decrease from this time. On the whole the data for the leg regeneration show a more extended period tlian do the tail regenerations. 132 ILLINOIS BIOLOGICAL MONOGRAPHS [132 TABLE 73 Amblystoma punctatum Series 4101-4540 Rate of regeneration per day of fore-leg at different times during the regener- ative process for three degrees of injury Period of regeneration Days Middle of period bays after operation Rate of regeneration per day for each period Average One fore-leg Both fore-legs Both fore- legs + one-half tail 0-2 1 0.06 0.08 0.07 0.07 2-4 3 0.04 0.10 0.07 0.07 4-6 5 0.10 0.08 0.13 0.10 6-8 7 0.12 0.15 0.09 0.12 8-10 9 0.12 0.25 0.25 0.21 10-12 11 0.28 0.18 0.18 0.21 12-14 13 0.25 0.29 0.34 0.29 14-16 15 0.52* 0.41* 0.39* 0.44* 16-19 17 Va 0.27 0.21 0.27 0.25 133] RATE OF RECEXERATIOS — ZELENY 133 TABLE 74 Amblystoma punctatum Series 4101-4540 Percentage increment per day of regenerating fore-leg at different periods for three degrees of injury Days Percentage increment per day during each period Average One fore-leg Two fore-legs Both fore- legs + one-half tail 2-4 34 62* 46* 47* 4-6 45* 23 45 38 6-8 28 28 16 24 8-10 19 30 35 28 10-12 31 9 15 18 12-14 17 18 21 19 14-16 26 19 17 21 16-19 14 10 13 12 Discussion The results obtained from the present study show that with certain material it is passible to control disturbing factors so as to get data of a sufficiently uniform nature for an analysis of the change in rate. Such material was found in the tails of the tadpoles of Bana clamitans. The analysis has yielded results which should be of value in a determination of the factors involved in the stimulation of growth and more particu- larly those concerned in slowing it down and finally bringing it to a stop. The characteristics of the change in rate have been studied by means of the curves of rate, of acceleration of rate and of percentage increments. Tlie rate is slow at first, increases rapidly until it is near a maximum at about eight days ; then decreases, at first rapidl.y and then more and more slowlj- as zero is approached. The acceleration of rate is plus only between the first two periods, i. e., up to the fifth day. After that it is minus, reaching its lowest point at ten days. The percentage increment 134 ILLINOIS BIOLOGICAL MONOGRAPHS [134 is very high between the first and second periods but decreases very rapidlj' at first and then more slowly. It is evident that there is a close similarity between the change in rate of growtli during the regeneration cycle and the change in rate during an ordinary developmental cycle and thei'e is every I'eason to believe tliat tlie factors controlling the one are similar to those controlling the other. The problem of the factors is particularly interesting when it is noted that for widely different levels the rates of regeneration differ in such a way that length regenerated in a given time is proportional to the length removed. The process of regeneration apparently is initiated in a similar manner at each level but is kept under such control that only a certain per cent of the length is regenerated in a given time. Knowledge of the process is at present insufficient to enable one to discuss with profit the nature of the control of rate of regeneration. All that can be done is to point out the relations of certain phenomena. The initial slow period is coincident with the period of cell migration without cell division, the period of rapidly increasing rate is coincident with the period of rapid cell multiplication without pronounced cell differentiation and the period of rapidly decreasing rate is associated with the appearance of pronounced differentiation in the cells. There is certainly some causal relation between these phenomena. SuMM.UtY 1. In second regenerations of the tail in Rana clamitans the average specific rates are 0.019 mm. for the to 4 day period, 0.066 for the 4 to 6 day period, 0.051 for 6 to 8 days, 0.033 for 8 to 10 days, 0.017 for 10 to 121/2 days, 0.001 for I21/2 to 18 days and —0.001 for 18 to 56 days. 2. Tlie average accelerations of rate are -|-0.095 mm. per day from the first to the second period, — 0.015 from the second to the third, —0.030 from the third to the fourth, —0.058 from the fourth to the fifth, —0.028 from the fifth to the sixth and —0.001 from the sixth to the seventh. 3. The average percentage increments between the same periods are respectively 106, 28, 12, 5, and 0. 4. The average accelerations of specific rate for the four deepest levels between the same periods are respectively +0.011 mm., 0.000, —0.005, —0.006, —0.004 and 0.000. 5. In first regenerations of the tail in Rana clamitans the average specific rates are 0.018 mm. for to 4 days, 0.046 for 4 to 6 days, 0.057 for 6 to 8 days, 0.037 for 8 to 10 days, 0.026 for 10 to 1214 days, 0.002 for 121/2 to 18 days and —0.001 for 18 to 56 days. 6. The average accelerations of rate are +0.07S mm. per day from 135] RATE OF REGESERATIOX —ZELESY 135 the first to the second period, 0.000 from the second to the third, — 0.022 from the third to the fourth, —0.042 from the fourth to the fifth, —0.025 from the fifth to the sixth and 0.000 from the sixth to the seventh. 7. The average accelerations of specific rate for the four deepest levels between the same periods are respectively +0.009, 0.000, — 0.004, —0.005, —0.003 and 0.000. 8. The average percentage increments between the same periods are respectively 98, 29, 17, 6, 1 and 0. 9. The experiments on salamander larvae show a similar change iu rate of regeneration during the process but the number of individuals is too small to allow an analysis of the data. 10. The changes in rate that have been noted bear a definite relation to the histological changes that have been observed during the regener- ation of the tail. 136 ILLINOIS BIOLOGICAL MONOGRAPHS [136 PART V THE EFFECT OF DEGREE OF INJURY UPON THE RATE OF REGENERATION In a former series of papers the writer gave the results of experi- ments on the effect of degree of injury upon the rate of regeneration. A number of different species of animals and various combinations of injuries were involved. The results then obtained tend on the whole to show that within certain limits the rate of regeneration from an injured surface is not retarded by simultaneous regeneration in other parts of the body. Where a difference exists between the rates with and without additional injury there is usually an advantage in favor of the part with additional injurj'. The differences are however often slight and in some of the cases come within tlie limits of probable error. It is only when the data as a whole are taken that it is possible to judge of the correctness of the general conclusion that within fairly wide limits of additional injury there is certainly no decrease in rate of regener- ation but rather a tendency toward an increase. Some additional data on these points have been obtained in connec- tion with the present study of the factors of regeneration. On the whole they confirm the previous results. The principal experiment (Experi- ment I) was planned with a view to further analysis of the problem, especially the determination of the effect of additional injury to a like organ as compared with additional injury to an unlike organ. Experiment I Amblystoma punctatum Series 4101-4540 The young were hatched on March 29-April 4, 1913, and the oper- ations were made on May 4 and 5. The measurements of the control individuals at the time of the operations are given in Table 75. The average total length is 31.3 mm., the tail length 14.4 mm., the average length of the fore-legs 3.6 mm. and the average of the hind-legs 1.5 mm. The measurements of control individuals at the end of the experi- ment on May 23 are given in Table 76. The total average length is 42.7 mm., the tail length 20.0, the average of the fore-legs 6.2 and the average of the hind-legs 4.5 mm. The experiment consisted in the determination of the regenerated length of the right fore-leg under three degrees of injury : when the 137] RATE OF REGEKERATIOX — ZELEXY 137 right fore-leg alone is removed, when its mate is also removed and finally when its mate and one-half of the tail are removed. In the last two eases the average of the two fore-legs is taken as the proper value for the regeneration 'of a fore-leg. A large number of individuals, all hatched from the same lot of eggs, were \ised and a selection of larvae was made so as to make the experimental animals as nearly alike as possible in this respect. In each of the five sets an individual for each degree of injury was kiUed at two days after the operation, and also at four, six, eight, ten, twelve, fourteen, sixteen and nineteen days. The data are given in Tables 77 to 88. The three degrees of injury may be represented by (1) R, (2) R-fL, (3) R-fL-fi/oT, in which R=right fore-leg removed, L^eft fore-leg removed and y2T=one-half of the tail removed. The second involves the removal of some additional material of the same kind as that removed in the first. The third as compared Avith the first involves the removal of some of the same kind of material and some of another kind. In every case it is the regeneration of the fore-leg that is used as the basis of comparison. The additional simultaneous injury and regeneration does not de- crease the regeneration of the individual fore-leg. At two days the average regenerated lengths of a fore-leg are respectively 0.13, 0.16 and 0.15 mm. for the three degrees of additional injury; at four days the corresponding values are 0.22, 0.36 and 0.29 ; at six days 0.42, 0.53 and 0.55; at eight days 0.66, 0.83 and 0.73; at ten days 0.91, 1.34 and 1.24; at twelve days 1.48, 1.60 and 1.61 ; at fourteen days 1.98, 2.19 and 2.29; at sixteen days 3.02, 3.01 and 3.08 ; at nineteen days 3.84, 3.64 and 3.90. At only two of the nine periods is the regeneration of the fore-leg witliout additional injury as rapid as that of a fore-leg with additional injury and at these two times it is less rapid than one of tlie two other groups. In seven of the nine cases the regeneration of the fore-leg witliout additional injury is less than either of the two with such injury. Among the fortj' individual comparisons in which all three degrees are present the degree with no additional injury has 6% firsts, the degree with an additional fore-leg 15% firsts and the degree with an additional fore-leg plus one-half of the tail has 17i/j firsts. Among the nine time groups the degree with no additional injury has IV-j firsts and eacli of the additional injury combinations has 3% firsts. Taking up the lowest positions in the three degrees in the same way, among the forty individual comparisons the degree with no additional injury gives the lowest regeneration in 2\\i\ cases while the additional injury combinations each have only 9i'-j lowest regenerations. Among the nine time groups the degree with no additional injury has the lowest value 6 times, the one with an additional removal of the other fore-leg 138 ILLIXOIS BIOLOGICAL MOXOGRAPHS [138 21/2 times while the one with the highest degree of injury gives tlie lowest regeneration for the fore-leg only i/o times. These comparisons show very clearly that the regeneration of a fore- leg is not as rapid when the individual is regenerating no other part at the same time as it is when the other fore-leg is being regenerated at the same time. The additional removal of one-half of the tail does not seem to accelerate the regeneration any further because there is no essen- tial difference between the effect of an additional injury of a fore-leg and an additional injury of a fore-leg plus one-half of the tail. It may be that the effect of additional removal is confined to removal of a similar part, the tail removal in this case involving a different kind of organ. Or it may be that the accelerating effect is found only within certain degrees of injury the limit being exceeding by the highest of the three degrees. Amblystoma punctatum Series 4101-4540 Experiment I Controls at beginning of experiment Date Cata- log number Total length mm. Tail length mm. Fore legs Hind legs Right Left Av'age. Right Left Av'age. 5/4/13 4110 35.0 16.4 4.0 4.0 4.0 3.0 3.1 3.05 5/4/13 4210 31.8 28.1 14.8 3.6 3.8 3.7 1.4 1.5 1.45 5/4/13 -4310 11.9 3.3 3.3 3.3 1.0 0.9 0.95 1 5/4/13 4320 33.8 15.3 13.8 3.3 3.1 3.2 1.1 1.0 1.05 J 5/5/13 4410 30.2 3.6 3.6 3.6 1.0 1.0 1.0 5/.?/13 4510 28.7 14.0 3.9 3.8 3.85 1.4 1.2 1.3 Average 31.3 14.4 3.6 1.5 139] RATE OF REGEXERATIOX—ZELEXY 139 TABLE 76 Amblystoma punctatum Series 4101-4540 Experiment I Controls at end of experiment Date Cata- log number Total length mm. Tail length mm. Fore legs Hind legs Right Left Av'age. Right Left Av'age. 5/23 4120 4130 4140 46.7 44.7 44.5 24.3 21.7 20.1 6.1 6.5 6.5 6.1 6.6 6.4 6.1 6.55 6.45 5.0 5.2 5.2 5.0 5.1 5.1 5.0 5.15 5.15 Average 45.3 22.0 6.4 5.1 5/23 4220 4230 4240 43.1 45.5 43.7 20.1 20.6 20.4 6.1 6.0 6.0 6.0 6.0 5.5 6.05 6.0 5.75 4.1 4.0 4.0 4.1 5.0 4.4 4.1 4.5 4.2 Average 44.1 20.4 5.9 4.3 5/23 4330 4340 47.2 41.5 21.9 19.5 7.1 6.1 7.2 6.2 7.15 6.15 5.3 4.0 5.2 4.1 5.25 4.05 Average 44.3 20.7 6.6 4.6 5/23 4420 4430 4440 41.0 40.5 40.4 18.2 19.4 18.9 7.0 5.6 6.0 7.0 5.6 6.0 7.0 5.6 6.0 4.9 4.0 4.1 4.8 4.1 4.0 4.85 4.05 4.05 Average 40.6 18.8 6.2 4.3 5/23 4520 4530 4540 36.5 40.9 40.6 39.3 16.0 18.5 19.1 5.6 6.7 5.0 5.6 6.8 5.0 5.6 6.75 5.0 4.0 4.9 4.0 4.0 4.8 4.0 4.0 4.85 4.0 Average 17.9 5.8 4.3 Grand average 42.7 20.0 6.2 4.5 140 ILLINOIS BIOLOGICAL MONOGRAPHS [140 TABLE 77 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injury Two days Catalog number 4101-11-21 4201-11-21 4301-11-21 4401-11-21 4501-11-21 Average Degree of injury One fore-leg 0.10 0.10 0.10 0.20 0.15* 0.13 Both fore-legs 0.22 0.15* 0.15 0.17 0.10 0.16* Both fore-legs -\- one-half tail 0.22 0.11 0.17* 0.20 0.07 0.15 TABLE 78 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injur} Four days Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs + one-half tail 4102-12-22 0.25 0.25 0.52 4202-12-22 — 0.52 0.22 4302-12-22 0.15 0.37* 0.22 4402-12-22 0.40* 0.30 0.27 4502-12-22 0.10 — 0.20 Average 0.22 0.36* 0.29 141] RATE OF REGENERATION— ZELENY TABLE 79 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injury Six days Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs + one-half tail 4103-13-23 0.40 0.20 0.92* 4203-13-23 0.50 0.87* 0.52 4303-13-23 0.45 0.65* 0.42 4403-13-23 0.45 0.60* 0.47 4503-13-33 0.30 0.35 0.42* Average 0.42 0.53 0.55* TABLE 80 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees o£ injury. Eight days Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs ■\- one-half tail 4104-14-24 0.50 0.75 0.97* 4204-14-24 0.80 0.80 0.80 4304-14-24 0.85 0.87* 0.62 4404-14-24 0.43 0.95* 0.75 4504-14-24 0.70 0.80* 0.52 Average 0.66 0.83* 0.89* 142 ILLINOIS BIOLOGICAL MONOGRAPHS [142 TABLE 81 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters tor different degrees of injury Ten days Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs -f one-half tail 4105-15-25 0.25 1.82* 1.60 4205-15-25 0.95 1.10* 1.07 4305-15-25 1.05 1.22 1.32* 4405-15-25 1.20 1.37* 1.07 4505-15-25 1.10 1.20* 1.12 Average 0.91 1.34* 1.24 TABLE 82 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injury Twelve days Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs -f one-half tail 4106-16-26 1.45 1.50 1.77* 4206-16-26 1.35 1.47* 1.44 4306-16-26 1.80* 1.60 1.65 4406-16-26 1.00 1.70* 1.50 4506-16-26 1.80* 1.72 1.67 Average 1.48 1.60 1.61* 143] RATE OF REGESERATION—ZELENY 143 TABLE 83 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injury Fourteen daj's Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs + one-half tail 4107-17-27 2.60 2.25 4207-17-27 1.70 1.97 2.22* 4307-17-27 1.45 1.87 1.95* 4407-17-27 2.25 2.72 2.90* 4507-17-27 1.90 2.12 Average 1.98 2.19 2.29* TABLE 84 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injury Sixteen days Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs -f one-half tail 4108-18-28 2.60 2.60 2.70* 4208-18-28 2.40 2.62* 2.22 4308-18-28 2.80 2.67 2.85* 4408-18-28 3.60 3.57 3.65* 4508-18-28 3.70 3.57 3.97* Average 3.02 3.01 3.08* 144 ILLINOIS BIOLOGICAL MONOGRAPHS [144 TABLE 85 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injury Nineteen days Degree of injury Both Catalog One Both fore-legs number fore-leg fore-legs + one-half tail 4109-19-29 4.00* 3.72 3.95 4209-19-29 3.65 4.05 4.25* 4309-19-29 3.60 2.85 3.60 4409-19-29 4.10* 3.95 3.80 Average 3.84 3.64 3.90* TABLE 86 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg in millimeters for different degrees of injury Summary Two to nineteen days Degree of Injury Days One fore-leg Both fore-legs Both fore-legs -f one-half tail 2 0.13 0.16* 0.15 4 0.22 0.36* 0.29 6 0.42 0.53 0.55* 8 0.66 0.83* 0.73 10 0.91 1.34* 1.24 12 1.48 1.60 1.61* 14 1.98 2.19 2.29* 16 3.02 3.01 3.08* 19 3.84 3.64 3.90* Groups first Groups last 7 4 2 5 145] RATE OF REGENERATION— ZELENY 145 TABLE 87 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg for different degrees of injury Tabulation of firsts for individual comparisons Injury Days One fore-leg Both fore-legs Both fore-legs -f one-half tail 2 1% iy2 2* 4 1 1 1 6 3* 2 8 Vi ^Vi* 1/3 10 4* 1 12 2 2 1 14 3* 16 1 4* 19 2 2 Total firsts i% 155/6 17/3 Groups first m 3^ 3^ 146 ILLINOIS BIOLOGICAL MONOGRAPHS TABLE 88 Amblystoma punctatum Series 4101-4540 Length of regenerated fore-leg for different degrees of injury Tabulation of lowest values for individual comparisons [146 Injury Days One fore-leg Both fore-legs Both fore-legs + one-half tail 2 3 1 1 4 1% 1% 1 6 3 1 1 8 2^ Yi 2^ 10 4 1 12 3 1 1 14 3 16 V2 3y2 1 19 1 2 1 Total lasts 21'^ 9^ 9'/3 Groups last 6 ZVz V2 Experiment II Amblystoma punctatum Series 4101-4540 This experiment deals with the same series of individuals as Experi- ment I. The comparison in this case however is one between the regen- eration of the removed half of the tail when it alone is removed and its regeneration when there is an additional removal of the two fore-legs. The data are given in Tables 89 to 99. At two days the regeneration of the tail without an additional injury is 0.35 mm. and with an additional injury 0.27. The corresponding values at 4 days are 0.73 and 0.81, at 6 days 1.32 and 1.31, at 8 days 2.06 and 2.26, at ten days 3.44 and 3.27, at twelve days 4.36 and 4.20, at fourteen days 4.82 and 4.94, at sixteen days 5.57 and 6.00 and at nineteen days 5.90 and 6.06. The regenerating tail with no additional injury is ahead at four times and the one with additional injury is ahead five times. In thirty three individual com- 147] RATE OF REGEX ERATION — ZELES Y 147 parisoQS the group with no additional injury is ahead seventeen times and the additional injury group sixteen times. Taking the individual cases by time groups the individuals with no additional injury are ahead 5^2 times and those with an additional injury 31/^ times. These comparisons show no advantage of one combination over the other. The additional removal of the fore-legs does not retard nor does it accelerate the regeneration of the tail. This result strengthens the view that the acceleration in Experiment I is probably due to the addi- tional removal of material similar to that whose rate is being studied. TABLE S9 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters for different degrees of injury Two days Catalog number 4131-21 4231-21 4331-21 4431-21 4531-21 Average Degree of injury One-half tail -f fore-legs 0.15 0.35 0.25 0.30 0.30* 0.27 TABLE 90 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters for different degrees of injury Four days Catalog number Degree of injury 148 ILLINOIS BIOLOGICAL MOXOGRAPHS [148 TABLE 91 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters tor different degrees of injury Six days Degree of injury Catalog number One-half tail One-half tail + fore-legs 4133-23 4233-23 4333-23 4433-23 4533-23 1.60 0.90 1.70* 1.10 1.00* 1.65 1.50* 1.10 Average 1.32* 1.31 TABLE 92 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters for different degrees of injury Eight days Degree 3f Injury Catalog number One-half tail One-half tail -\- fore-legs 4134-24 4234-24 4334-24 4434-24 4534-24 2.40 1.80 1.80 2.70* 1.60 2.60* 1.90* 2.26* 2.30 Average 2.06 2.26* 149] RATE OF REGESERATION—ZELENY 149 TABLE 93 Amblystoma punctatum Series 4101-4540 Lengtli of regenerated tail in millimeters for different degrees of injury Ten days Degree of injury Catalog One-balf One-half number tail tail -f fore-legs 4135-25 3.65* 3.20 4235-25 2.55 4335-25 3.20 1.46 4435-25 3.65* 3.20 4535-25 3.25 4.15* Average | 3.44* 3.27 TABLE 94 Amblystoma punctatum Series 4101-4540 Lengtli of regenerated tail in millimeters for different degrees of injury Twelve days Catalog number 4136-26 4236-26 4336-26 4436-26 4536-26 Average Degree of injury One-half tail fore-legs 150 ILLIXOIS BIOLOGICAL MOXOGRAPHS [ISO TABLE 95 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters for different degrees of injury Fourteen days Degree of injury Catalog number One-half tail One-half tail + fore-legs 4137-27 4237-27 4337-27 4437-27 4537-27 4.80 4.90* 5.00* 4.90* 4.50 6.00* 4.70 4.95 4.00 5.05* Average 4.82 4.94* TABLE 96 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters for different degrees of Injury Sixteen days Degree of injury Catalog number One-half tail One-half tail -j- fore-legs 4138-28 4238-28 4338-28 4438-28 4538-28 6.50* 5.80 5.00 5.00 8.00 5.50 6.40* 6.10 Average 5.57 6.00* 151] RATE OF RECEXERATIOX — ZELENY 151 TABLE 97 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters for different degrees of injury Nineteen days Degree Df injury Catalog number One-half tail One-half tail 4139-29 4239-29 4339-29 4439-29 6.90* 3.20 4.90 5.90 6.20 5.55 6.60* Average 5.90 6.06* TABLE 98 Amblystoma punctatum Series 4101-4540 Length of regenerated tail in millimeters for different degrees of injury Summary Two to nineteen days Degree Df injury Days One-half tail One-half tail + fore-legs 2 0.35* 0.27 4 0.73 0.81* 6 1.32* 1.31 8 2.06 2.26* 10 3.44* 3.27 12 4.36* 4.20 14 4.82 4.94* 16 5.57 6.00* 19 5.90 6.06* Groups first 4 5 152 ILLINOIS BIOLOGICAL MONOGRAPHS [152 TABLE 99 Amblystoma punctatum Series 4101-4540 Length of regenerated tail for different degrees of injury Tabulation of firsts for individual comparisons Injury Days One-half One-half tail tail + fore-legs 2 21/2* iy2 4 21/2 21/2 6 1 2* 8 1 3* 10 2* 1 12 3* 2 14 3* 2 16 1 1 19 1 1 Total firsts 17 16 Groups first 51/2 314 Experiment III Amblystoma punctatum Series 4005-4008 Experiments III, IV, V and VI comprise merely a few individual comparisons obtained from experiments devised principally for the study of other factors. They are included here under the rule that no valid data on the matter at hand are to be excluded. In Experiment III the regeneration of the hind-leg is compared under the four conditions of (1) no additional injury, (2) removal of the other hind-leg, (3) removal of the other hind-leg and one fore-leg and (4) removal of the other hind-leg and both fore-legs. The data are given in Table 100. Three sets of comparisons were made at twelve days after the oper- ations, each with a single individual for each degree of injury. The regenerating hind-leg with no additional injury is distinctly behind the cases with additional injury. The greatest regenerated length comes in one case with an additional injury of one hind-leg plus one fore-leg and in two cases with one hind-leg plus two fore-legs. The averages begin- 153] RATE OF RECESERATIOS — ZELENY 153 niiig with the lowest degree of injury are respectively 1.50, 1.73, 1.86 and 1.88 mm. The additional removals are in every case removals of leg material and the result agrees with that of experiment I in giving an increased rate of regeneration of a part when similar organs are removed at the same time. TABLE 100 Amblystoma punctatum Series 4005-4008 Length of regenerated hind leg in millimeters for different degrees of injury Twelve days Catalog number Degree of injury One hind-leg Both hind-legs Both hind- legs+one fore-leg Both hind- legs -f both fore-legs 4005 4006 4008 1.35 1.65 1.50 1.90 1.80 1.50 1.95* 1.82 1.80 1.75 1.92* 1.85* Average 1.50 1.73 1.86* 1.84 TABLE 101 Amblystoma punctatum Series 400.')-4008 Length of regenerated fore-leg in millimeters for different degrees of injury Twelve days Degree of injury Catalog number One Both fore-leg fore-legs + both + both hind-legs hind-legs 4005 3.0* 2.8 4006 3.1* 3.0 4008 3.0 3.15* Average 3.07* 2.98 Experiment I\' A.mulystoma punctatum Series 4005-4008 In this expeiiiiicnt tiie regeneration of the right fore-leg is compared under conditions of differing degrees of additional injury. In one com- bination there is an additional removal of the two hind legs and in the 154 ILLINOIS BIOLOGICAL MONOGRAPHS [1S4 otlier of both hiud-legs plus the remaining fore-leg. The data are given in Table 101. In two of the three cases the smaller additional degree of injury shows the greater regeneration of the fore-leg. The average is 3.07 mm. for the lesser degree and 2.98 for the greater degree, an advantage in favor of the lesser degree. It should be noted that this is not strictly comparable with the main issue of Experiments I, II and III. Aside from tlie small number of cases it is a comparison between two degrees of injury each of which is of considerable extent. It may be that the removal of three of the four legs is near the degree of injury yielding the maximum rate for each removed leg. Experiment V Amblystojia punctatum Series 4010-4025 A comparison is made between the regeneration of a half of the tail when it alone is removed and when both fore-legs are removed at the same time. Four individual comparisons are made at fourteen days. The data are given in Table 102. The regenerated lengths and specific lengths regenerated are ahead in two of the four cases for each of the degrees of injury. The average regenerated length with no additional injury is 5.1 mm. and with additional injury 5.0 mm. The specific regenerated length is 0.65 with no additional injury and 0.68 with addi- TABLE 102 Amblystoma punctatum Scries 4010-4025 Regeneration of tail for different degrees of injury Fourteen days Degree of injury C )ne-half tai Length regener- ated One-half tail -|- both fore-legs Catalog number Length removed Specific amt. re- generated Length removed Length regener- ated Specific amt. re- generated 4014-13 7.7 4.9 0.64 7.0 5.2* 4.3 0.74* 4018-17 8.8 5.2* 0.59* 8.0 0.54 4022-21 8.0 5.3* 0.66* 8.0 5.1 0.64 4025-24 7.0 4.9 0.70 6.6 5.3* 0.80* Average 5.1 0.65 5.0 0.68 155] RATE OF REGENERATIOX—ZELENY 155 tioual iujurj'. The data show essential equality between the rates of regeneration under the two conditions of the experiment. This agrees with the data in Experiments I and II which show no increase or decrease in rate of regeneration when unlike material is removed simultaneously with the removal of the organ whose rate is being studied. Experiment VI Ambltstoma punctatuii Series 4010-4025 Three individual comparisons were made at fourteen days of the right fore-leg, when it alone is removed, when the other fore-leg is also removed and when the other fore-leg plus one half of the tail is removed. The data are given in Table 103. In two of the three cases the individuals TABLE 103 Amblystoma punctatum Series 4010-4025 Length of regenerated fore-leg in millimeters for different degrees of injury Fourteen days Degree of injury Catalog number One fore-leg Both fore-legs Both fore-legs + one-half tail 4011,12,13 4015, 16, 17 4019, 20, 21 4023,— ,24 2.00* 2.00* 1.95 2.00 1.77 1.60 1.82 1.65 1.80 2.22* 2.00 Average 1.99* 1.73 1.92 with no additional regeneration are ahead of the others. The greater injury gives the greater rate in one of the three. The average regener- ated lengths beginning with the lowest degree of injury are respectively 1.99, 1.73 and 1.92 mm. The few cases may be a sufficient explanation of the lack of agreement with the more extended series of Experiment I. Discussion The experiments as a whole show that a part regenerates slightly more rapidly when additional material of the same kind is removed tlian when the part alone is removed. Simultaneous removal of tail material does not accelerate the regeneration of a leg nor does simultaneous re- moval of a leg accelerate the regeneration of the tail. The rate in these eases however is not decreased by the additional injurj'. The state- 156 ILLINOIS BIOLOGICAL MONOGRAPHS [156 ment may therefore be made that within limits the regeneration of a part is not retarded by simultaneous removal and regeneration of material in other parts of the body. When this additional material is of the same kind as that whose rate is being studied there may even be an acceler- ation of regeneration. In comparison with such a factor as level of the cut this difference in rate is slight and no such quantitative relation as in that case can be made out. It must however be considered that the principal object of the original experiments was to show that additional injury within the given limits tends to increase rather than decrease the rate of regeneration. This has been proved for these experiments. The evidence in favor of a definite increase in rate with any certain increase in degree of injury is not so conclusive. It is obvious that in many series of experiments factors whose influence is greater than that of the factor under discus- sion may obscure the result. Emphasis should again be placed on the fact that all data obtained by the writer are included. That some of the series, especially those with a few individuals, diverge from the general result is to be expected by anyone in similar work who has attempted to eliminate entirely all of the factors except the one under observation at a particular time. Summary 1. A comparison was made of the rate of regeneration of a leg or of the tail of an Amblystoraa larva when the part alone is removed with its rate when similar or dissimilar parts of the individual are removed at the same time. The data are derived from two principal Experiments, I and II, and from a few scattered observations listed as Experiments III to VI. 2. In Experiment I a comparison was made of the rate of regener- ation of the right fore-leg when it alone is removed witli its rate when the other fore-leg is removed at the same time and when the other fore- leg and one half of the tail are removed. The result obtained from forty individual comparisons made at different times shows that the rate of regeneration of the right fore-leg in each of the series with additional injury is greater than in the series without additional injui-y. 3. The rate of regeneration of a right fore-leg when its mate plus one-half of the tail is removed is not essentially different from the rate when its mate alone is removed. The addition of the injury to a dissimilar organ, the tail, does not alter the rate of regeneration of the fore-legs. 4. In Experiment II it is shown that there is no significant differ- ence between the rate of regeneration of a tail one-half of which has been 157] RATE OF REGEXERATIOX — ZELEXY 157 removed without additional injury to the individual and the rate after the same injury plus a removal of both fore-legs. 5. The data of Experiments III to VI show some departures from the general rule probably because they deal with few individuals. On the whole however they bear out the results obtained from the principal experiments. 1S8 ILLINOIS BIOLOGICAL MONOGRAPHS [158 PART VI THE COMPLETENESS OP REGENERATION One of tlie striking facts in connection with amphibian regeneration as made out in the present studies is the lack of completeness of the process. When a part of the tail is removed the lost part is never com- pletely restored. Data on this problem are to be found in a number of sets of experiments one of which (Experiment V) was devised especially for the present purpose. Experiment I Rana clamitans Series 3557-3624 One-half of the tail was removed in the individuals of three groups, A, B and C. After 35 to 39 days, which was sufficiently long so that regeneration had stopped, another removal was made and so on until eaeli individual had undergone five regenerations. The data are given in Table 104. The aA'erage removed length as estimated from the measure- ment of a few individuals was 17.0 mm. The average length of the com- pleted first regeneration is 8.6 mm. or 51 per cent of the removed length, of the second regeneration 8.0 mm. or 53 per cent, of the third 7.5 mm. or 51 per cent, of the fourth 5.5 mm. or 42 per cent and of the fifth 6.4 mm. of 45 per cent. On the average about one-half of the removed length is replaced when one-half of the tail length is removed. Experiment II Rana clamitans Series 3628-3675 One-half of the tail length was removed in the individuals of this experiment and regeneration was allowed to proceed for twenty days, a sufficient time for bringing it to a stop. The data are given in Table 105. The average original tail .length was 21.8 mm., of the removed length 10.6 mm. and of the regenerated length 5.4 mm. The completed regenerated length is thus 51 per cent of the removed length. Experiment III Rana clamitans First regenerations Series 3676-3765 The data are given in Table 106. The tails were removed at different levels approximating 6, 10, 17, 30, 48 and 62 per cent of the tail lengths. Regeneration was completed at these levels at 121/2, 121/^, 12i/^, IS, 18 159] RATE OF REGEXERATIOX—ZELEXV 159 and 56 days respeetivel.y. The regenerated lengths at these times of completion are respectively 61, 46, 39, 33, 42 and 41 per cent of the removed lengths. It will be noted that the two shortest removals give the highest per cents and the two medium ones the lowest per cents. This difference is discussed in Part III on the effect of level of the cut. Experiment IV Rana clahitaxs Second regenerations Series 3676-3765 The data are given in Table 107. The tail was removed at different levels approximating 6, 10, 18, 31, 49 and 67 per cent of the removed lengths. Regeneration was completed for these levels at 10, 10, 12i/^, TABLE 104 Rana clamltans Series 3557-3624 Completeness of regeneration Successive regenerations in single individuals One-half of tail removed = 17 mm. on the average First operation Oct. 23. 1911 Second operation Groups A and B Nov. 18 Group C Nov. 28 Group A Group B Catalog number First regener- ation Nov. 28 3564 3565 3566 3567 3568 3569 3570 Average 3578 3579 3580 3581 3582 3583 3584 Average Second regener- ation Jan. 3 9.5 9.8 10.0 11.9 8.4 10.0 8.7 9.8 8.3 8.2 11.9 9.7 8.3 Third regener- ation Feb. 9 8.5 11.4 9.3 9.5 9.9 8.7 8.1 9.3 9.0 8.1 7.3 8.0 12.8 7.4 9.5 8.9 Fourth regener- 7.3 8.1 8.0 11.0 8.0 8.5 5.8 11.3 6.9 7.6 7.4 5.0 6.4 7.2 Fifth regener- April 24 8.2 6.3 11.9 8.5 8.1 8.6 8.2 7.0 8.1 11.6 5.4 6.8 160 ILLIXOIS BIOLOGICAL MONOGRAPHS (160 TABLE 104 (Coutinued) First Second Third Fourth Fifth regener- regener- regener- regener- regener- Catalog ation ation ation ation ation number Nov. 28 Jan. 3 Feb. 9 Mar. 16 April 24 3586 8.1 — 3588 6.1 7.5 7.3 5.7 7.2 3590 8.5 6.6 7.5 5.1 6.5 3592 7.1 8.0 5.7 4.9 6.5 3594 8.6 7.8 2.0 5.0 6.1 3596 9.0 8.6 9.4 6.7 6.8 3598 10.7 9.7 9.3 6.1 6.2 3600 8.2 8.0 6.9 5.8 5.9 3602 9.9 7.7 6.8 4.4 4.7 Group 3604 9.6 7.6 6.C 4.9 5.7 C 3606 7.4 7.8 8.0 5.0 5.9 3608 9.0 8.0 9.0 5.5 6.9 3610 8.5 8.9 8.3 5.4 7.1 3612 7.4 7.0 7.1 4.8 5.5 3614 8.3 6.6 6.2 4.5 5.2 3616 8.0 8.3 7.9 6.1 7.9 3618 9.7 9.5 9.7 7.3 8.0 3619 — 8.0 7.3 6.6 6.1 3622 10.2 8.5 — — — 3624 9.3 7.5 — — — Average 8.6 8.0 7.5 5.5 6.4 Percent of removed length r egen. Av. 51 53 51 42 45 121/2, 56 aiul 56 days respectively. The regenerated lengths at these times of completion are respectively 67, 46, 33, 31, 40 and 39 per cent of the removed lengths. As in the case of the first regenerations the two shortest removals give the highest per cent of regeneration and the two medium removals the lowest per cent. Experiment V Amblystoma punctatum Series 6212-6281 The experiments on tadpoles of Rana clamitans having shown that only a half or less of the removed length on the average is completed dur- ing regeneration it became a matter of interest to see if this might not have been due to the age of the tadpoles, which were obtained in the fall. Accordingly a series of Amblystoma larvae was operated upon within a 161] RATE OF REGENERATION —ZELENY 161 few days after they had left the egg envelopes and was kept until the salamanders were weU advanced in their metamorphosis. Since in young salamander larvae the border line between old and regenerated tissue is soon obliterated it became necessary to devise another method of testing completeness of regeneration than the direct measurement of the regen- TABLE 105 Rana clamitans Series 3628-3675 Tail Removed length length 24.1 13.1 24.6 13.2 22.1 11.0 23.2 11.1 23.1 11.7 25.0 12.5 20.4 9.9 20.8 10.0 29.2 15.5 23.8 10.5 23.3 10.9 25.6 10.9 20.8 10.1 19.2 9.8 21.1 11.5 22.0 11.8 17.0 8.2 19.0 9.7 22.4 9.8 19.8 10.1 20.8 8.4 21.8 9.1 15.4 7.3 18.1 9.0 21.8 10.6 Percent removed 49 Regenerated length Twenty days 5.9 5.2 4.9 5.5 5.9 5.8 5.6 5.2 5.9 5.6 5.6 5.9 4.7 4.6 5.5 6.1 5.6 5.5 4.3 5.2 5.1 5.0 6.0 6.0 Percent of removed part regenerated 162 ILLINOIS BIOLOGICAL MONOGRAPHS [162 TABLE 106 Rana clamitans Series 3676-3765 First regenerations Average Percent Tail maximum Average Days after Number of tail length regeneration maximum operation of length removed in regeneration when maxi- cases r emoved in mm. percent of in mm. mum is Average Average removed length reached Average 2 6 10 1.5 61 0.9 12y2 5 2.6 46 1.2 12 Va 3 17 4.6 39 1.8 12 Vo 8 30 8.2 33 2.7 18 5 48 13.0 42 5.5 18 5 62 16.7 41 6.9 56 TABLE 107 Rana clamitans Series 3676-3765 Second regenerations Average Percent Tail maximum Average Days after Number of tail length regeneration maximum operation of length removed in regeneration when maxi- cases removed in mm. percent of in mm. mum is Average Average removed reached length Average 4 6 1.5 67 1.0 10 7 10 2.8 46 1.3 10 5 18 4.9 33 1.6 12 Vs 10 31 8.4 31 2.6 121/2 8 49 13.1 40 5.2 56 10 67 18.1 39 7.1 56 163] RATE OF REGEXERATIOX —ZELEKY 163 erated material. This consisted iu a comparison of the ratio between tail length and body length in the operated individuals with that in control iinoperated individuals. This was done after regeneration had been tail going on during the whole larval period. If tlie — — period is the same in operated as in unoperated individuals it is proper to suppose that regeneration has been complete. If however the ratio is lower the conclusion that regeneration is incomplete is very probably correct though absolute certainty can not be assumed because of the possibility of the changed ratio being due to regulatory changes in other parts of the indi\adual. The experiment consists of a comparison of the relative degree of completeness of regeneration of the tail in four groups, (1) with no operation, (2) with one-fourth of the tail removed, (3) with one-half of the tail removed and (4) with three-fourths removed. The operations ■were made as soon as possible after the animals left the egg envelopes and the experiment proceeded until all four legs were well developed and absorption of the gills had begun. This allowed practically the entire larval period for regeneration. There were seventy individuals at the start but a high mortality reduced the number very considerably. Lim- nodrilus was used as food. The data are given in Tables 108 to 112. The average ratio between tail and body length in control individuals at the end of the experiment is 1.09, in indi\-iduals with one-fourth of the tail removed it is 1.01, in those with one-half removed 0.93 and with three-fourths removed 0.S6. This progressive relative decrease iu the tail length as compared with the body length is very probably due to lack of completeness of regeneration even though the whole larval period has been allowed for such completion. Discussion Apart from the starting stimulus in regeneration the most interesting problem is undoubtedly that of the stopping stimulus. With the growth once started what are the factors involved in checking it? In general it has been assumed that regeneration goes on until the removed organ is entirely replaced and that over- and under-regeneration occur but rarely. The present data make it probable that incompleteness is more general than has been supposed. The factors at work iu bringing regeneration to a close tend to overdo rather than underdo their function. A further investigation of the problem of completeness of regener- ation would be of interest. 164 ILLINOIS BIOLOGICAL MONOGRAPHS [164 >> -a e3 03 o a> H m ^MOSOOOOlrtO^ o> a o Is OrHOr^t-lOrHrHO o T- OJ V, O S A >.5 COOK^OO^^OOC ■* "a tm Cv3C>ilOU2-^Lr300*<*^lO -COt-^t^I>;oq I^ o. 6 OOOOOOOOO o Y, CS t»-i K o S "^ >> 5 o S 00l>;t-COCOCOCa:DCO CO (J a < ' 165] RATE OF REGENERATIOS —ZELEKY 165 1 a ___, >. ,_, 00 -^ M •0 oa 05 q a eg ^ d rH (D lO >. J3 C<1 CO ^ to O 5a ■a o a M iii M -W ■*J^' •* ■O tH 3 "^ '"* ■^ 3 O* 1 •a H 3 3 o B i! T «u. « lO 5 l-j *rf tD 00 -^ CO 'T J) to E- 3 T-1 '-' ■^ '3 c "" 1 _ o •a S "3 X3 ea L-; t-; CO 5 5 O Eh in a 0) CO to CX3 t^ M 00 3 O ■^ a P ' O >• t- cS C CO 00 £ — d d d d d ^ 2^ a a I 1 pa 2 V ^ ^ t- to i^ s Ml 3 r-i rt ,H « ■r^ < ,r X V _a) Eh J 0) tf o s •a CO M s to oa to 00 3 0a ■3 Eh t- 00 t- 1^ s 3 tH d d d d 6 3 oa 3 O c 3 '5b iH m 3. >. ■3 eo to •«• m o> 06 I-; B. J3 « :d s o a ja ooa ^ >> tD ■5 to 3 to CD LA CO to 2 a E- a '> -* iH rH tH " b 5 ■ 3 < o ^ 1 1 3 a ^ >> ". '^. . 0) ^ 1-^ '5 to ■^ ^ •^ ^ cvi ^ M) E- c tH r-i rH ^~ f;;^ C .^ '3 0) •rj _2 "3 J3 in ^ t- (> CM CM :7^ M) 00 CO 00 •<** 10 j= ^ 3 M N M CM 6 3 ^^^^ •a V ■* t- oa • « 10 0, a to 3 M co' ed CO CO 0) X , OJ 2 < •0 n 50 3 00 00 t> 00 00 a ^ 3 ^ „ ft CO oa 00 oa to a •< bo 3 to to to to t^ to 3 > h 3 _2 ] hJ s g CO eo_ to •^ h- Eh Ml 3 1-i V S & M= _« !lO 5 0) a i in m ■a- « 01 n 0! 3 to to to to > u 3 < 166 ILLIXOIS BIOLOGICAL MONOGRAPHS [166 11 a a © __ ■a o M t- t- (O '3 CTi ■^ CO 00 H ra "= o d d X ^ J3 M ^ CO to ° S ■a o pq OJ c\i •a r-( ^ '"' ■^ H S ; t- xt< rv. a ■— O o t^ d ^_, 9 S^ '3 00 t-; t- r^ a '-' o O d d a oi M '-' 01 _ m -s ^ a >. ;^ IT- !>; o 00 < ■a o t-^ O^ o6 N. 2 K o M c: _a; „ ^ -* T-t tH w 60 '3 tj CO CO CO to H Si '> J ^ si c« r-\ o -*" oo ■^' ■^ '-I — be o u ~~^ CO ^ CO "3 a ■^ -S OS 3 CD CO CD > o c < H W O O C> CO •r^ T^ d d CO OJ CM CJ £ .- r^ r-- 00 r^ o d d o i •? -? 167] RATE OF REGEXERATIOX—ZELENY 167 BIBLIOGRAPHY Abel M. ig02. Beitrage zur Kenntnis der Regenerationsvorgiinge bei den limicolen Oligochaeten. Z. f. wiss. zoo!., 63:1-74. Allen, W. E. 191 1. A Study of the Relation of Tissue Differentiation to Rate of Growth during Regeneration. Biol Bull., 21 : 187-206. Barfurth, D. 1903. Die Erscheinungen der Regeneration bei Wirbeltierembryonen. O. Hertwig's Handbuch der Entwickelungslehre der Wirbeltiere. Bonnet, C. 1745. Traite d'insectologie. Seconde partie. Observations sur quelques especes de vers d'eau douce, qui, coupe par morceaux deviennent autant d'animaux complets. Paris. Davenport, C. B. 1899. Experimental Morphology, Part II, New York. Driesch, Hans 1897. Studien iiber das Regulationsvermogen der Organismen, I. Von den regulativen Wachstums-und Differenzirungs-fahigkeiten der Tubularia. Arch. f. Entw. Mech., 5 :389-4i8. DuRBiN, Marion L. 1909. An Analysis of the Rate of Regeneration Throughout the Regener- ative Process. J. Exp. Zool., 7 1397-420. Ellis, M. M. 1909. The Relation of the Amount of Tail Regenerated to the Amount Removed in Tadpoles of Rana clamitans. J. Exp. Zool., 7 :42l-456. Em MEL, V. E. 1906. The Relation of Regeneration to the Molting Process in the Lobster. Thirty-sixth Annual Report of the Commissioners of Inland-Fisheries of Rhode Island. Special paper, no. 27:257-313. Kammerer, Paul 1905. L'eber die Abhangigkeit des Regenerationsvermogens der Amphib- ienlarven von Alter, Entwicklungsstadium und spezifischer Grosze. Arch, f. Entw. Mech., 19:148-180. King, Helen D. 1898. Regeneration in Asterias vulgaris. Arch. f. Entw. Mech., 7:351-363. MiNOT, C. S. 1908. .Age, Growth and Death. New York. 168 ILLINOIS BIOLOGICAL MONOGRAPHS [168 Morgan, T. H. 1902. Further Experiments on the Regeneration of the Tail of Fishes. Arch, f. Entw. Mech., 14:539-561. 1906. The Physiology of Regeneration. J. Exp. Zool., 3 :457-50O. 1909. The Dynamic Factor in Regeneration. Biol. Bull., 16 :26s-276. MORGULIS, S. 1907. Observations and Experiments on Regeneration in Lumbriculus. J. Exp. Zool., 4:549-574. igoga. Regeneration in the Brittle-Star Ophiocoma pumila, with Special Ref- erence to the Influence of the Nervous System. Proc. Amer. Acad, of Arts and Sc, 44:655-659. 1909b. Contributions to the Physiology of Regeneration. I. Experiments on Podarke obscura. J. Exp. Zool., 7 :595-642. 1909c. Contributions to the Physiology of Regeneration. II. Experiments on Lumbriculus. Arch. f. Entw. Mech., 28:396-439. Przibram, Hans 1906. Aufzucht, Farbwechsel und Regeneration einer agyptischen Gottesan- beterin (Sphodromantis bioculata Burm). Arch. f. Entw. Mech., 22: 149-192. Scott, G. G. 1907. Further Notes on the Regeneration of the Fins of Fundulus heterocli- tus. Biol. Bull., 12:385-400. 1909. Regeneration in Fundulus and its Relation to the Size of the Fish. Biol. Bull., 17:343-353. Spallanzani, Lazaro Abbe 1769. An Essay on Animal Reproductions. Translated by M. Maty. London. Stockard, C. R. 1908. Studies of Tissue Growth, I. ./^n Experimental Study of the Rate of Regeneration in Cassiopea xamachana. Carnegie Institution Publication No. 103 :63-i02. 1909a. Studies of Tissue Growth, II. Functional Activity, Form Regulation, Level of the Cut, and Degree of Injury as Factors in Determining the Rate of Regeneration. The Reaction of Regenerating Tissue in the Old Body. J. Exp. Zool., 6:433--i7i. 1909b. Studies of Tissue Growth. IV. The Influence of Regenerating Tissue on the Animal Body. Arch. f. Entw. Mech., 29:15-32. Ubisch, Leopold v, 1915. Uber den Einflusz von Gleichgewichtsstorungen auf die Regenerations- geschwindigkeit. Arch. f. Entw. Mech., 41 :237-2SO. Vanlair, C. 1894. Recherches chronometriques sur la regeneration des nerfs. .Archives de physiologic normale et pathologique. 5^ Serie., 6:217-231. Zeleny, C. 1902. A Case of Compensatory Regulation in the Regeneration of Hydroides dianthus. Arch. f. Entw. Mech., 13 ;597-6o9. 1903. A Study of the Rate of Regeneration of the .Arms in the Brittle-Star, Ophioglypha lacertosa. Biol. Bull., 6:12-17. 169] RATE OF REGEXERATIOX—ZELENY 169 iposa. Compensatory Regulation. J. Exp. Zool., 2:1-102. iposb. The Relation of the Degree of Injury to the Rate of Regeneration. J. Exp. Zool., 2:347-369. 1907. The Effect of Degree of Injury, Successive Injury and Functional Activity upon Regeneration in the Scyphomedusan, Cassiopea xamachana. J. Exp. Zool., 5:265-273. 1908. Some Internal Factors Concerned with the Regeneration of the Chelae of the Gulf-Weed Crab (Portunus sayi). Carnegie Institution Publica- tion No. 103:103-138. 1909a. The Effect of Successive Removal upon the Rate of Regeneration. J. Exp. Zool., 7:477-512. 1909b. The Relation between Degree of Injury and Rate of Regeneration — Additional Observations and General Discussion. J. Exp. Zool., 7 :5i3-562. 1909c. Some Experiments on the Effect of Age upon the Rate of Regener- ation. J. Exp. Zool., 7:563-593. 1912. The Quantitative Study of the Internal Factors Controlling Regenera- tion. Proc. Seventh Intern. Zool. Congress 1907 :49i-494. Zuelzer, M. 1907. Uber den Einflusz der Regeneration auf die Wachstumsgeschwindig- keit von .-Xsellus aquaticus. Arch. f. Entw. Mech.. 25 :36i-397. ILLINOIS BIOLOGICAL MONOGRAPHS Vol. Ill October, 1916 No. 2 Editorial Committki: Stephen Alfred Forbes William Trelease Henry Baldwin Ward Published under the Auspices of the Graduate School nv THE University of Illinois Coi'YRlGHT, I915 By the Univkrsity of Illinois Distributed December 30, 1916 THE HEAD-CAPSULE AND MOUTH-PARTS OF DIPTERA WITH TWENTY-FIVE PLATES ALVAH PETERSON Contributions from the Entomological Laboratories of the University of Illinois No. 52 THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Entomology in the Graduate School of the University of Illinois 1915 TABLE OF CONTENTS PACE Introduction 7 Methods 8 Acknowledgments 9 Materials 9 Fixed Parts of the Head 13 Epicranial Suture 14 Fronto-clypeus 17 Tormae ..._ 19 Ptilinum 20 Labrum 20 Vertex 21 Compound Eyes and Ocelli 22 Occiput and Postgenae ; 23 Tentorium 26 iSIovable Parts of the Head 32 Antennae : 33 Mandibles 34 Maxillae 36 Labium 4' Epipharynx and Hypopharynx 49 Summary 54 Bibliography 57 Explanation of Plates 6i 177] HEAD OF DIPTERA — PETERSOX INTRODUCTION The head and mouth-parts of Diptera offer a rich field for research. A number of excellent studies have been made bj' several investigators and the}' deserve careful consideration. A review of practically all the literature shows that a majority of the workers have examined only one or a few species. Meinert (1881) and Hansen (1883), however, studied a number of forms, bvit they were mostly specialized species; wliile an important study by Kellogg (1899) deals only with the families of the Nematocera. Becher (1882) is the only investigator who has studied a large series of generalized and specialized species. I have made a special effort to secure as many generalized and specialized species as possible, since it is highly desirable and essential in homologizing structures to have at hand a wide range of species. Extensive studies have not heretofore been made, so far as I know, on the head-capsule; consequently the important relationship which ex- ists between the mouth-parts and the head-capsule in generalized insects has not been traced in Diptera. This relationship is just as significant in ascertaining the correct interpretation of the mouth-parts of Diptera as it is in other orders. Its importance is illustrated by a study of the head and mouth-parts of the Thysanoptera (Peterson, 1915). A review of the literature, Dimmock (1881) or Hansen (1883), discloses the many and varied interpretations that have been given to the mouth-parts of Diptera. To arrive at a correct interpretation of the fixed and movable parts of the head, the head-capsule and mouth-parts of all the species studied, irrespective of the established systematic position of the species, have been carefully compared with the' head and mouth-parts of generalized insects. On the basis of this comparison, generalized, hypothetical types have been constructed for each fixed and movable part. Each hypothetical type is made up by an accumu- lation of all the generalized characters found among the Diptera, and should show an intermediate stage between generalized insects and Dip- tera. The use of such a hypothetical type is a great aid not only in showing how the dipterous type has been developed, but also in deter- mining the homology of the parts. The scope of this investigation makes it necessary to limit the dis- <;ussions to the general subject of homology; consequently many details 8 ILLINOIS BIOLOGICAL MONOGRAPHS [17fr of structure and other interesting modifications, shown in the figures but without direct bearing on the subject of homology, are necessarily disregarded. The fixed and various movable parts of the head are dis- cussed separately, as developed from the hyjiothetical types, the discus- sions in every case proceeding from the generalized to the specialized. All the general conclusions pertaining to the head and mouth-parts presented in the following pages are based entirely on a study of the species listed under "materials", unless otherwise stated. General statements in respect to the mouth-parts are true only for species having them well developed. The names here adopted for the sclerites of the head and mouth- parts have been made to agree, so far as possible, with the terms now in common use for the same parts in generalized insects. The terms most commonly used thruout the literature for structures peculiar to this order have been adopted unless clearly unsuitable ; and new terms have been applied only to structures described here for the first time and to parts to which the current names are inappropriate. METHODS The greater part of this study was made from dried specimens that had been soaked from two to twenty-four hours in a 10% solution of potassium hydroxide. The sclerites of weakly chitinized forms show more clearly when they have been soaked for only a short time. After soaking, the heads were washed in distilled water to remove the potas- sium h.ydroxide and then preserved in 70% alcohol. All dissections were made under a binocular microscope in 70% alcohol in deep watch-glasses or in carbol-aniline oil. Studies and figures were largely made from dissected parts in alcohol. Cleared preparations moiinted in balsam were also found useful. In making such preparations the parts were dissected, stained, and cleared in carbol-aniline oil. This oil evaporates slowly, will mix readily with safranin or orange G dissolved in 95% alcohol, and will clear from any grade of alcohol above 50%. Tlie staining of material with safranin before mounting proved to be very useful in differentiating the almost colorless parts of some species. When using aniline oil it is necessary to remove as much as possible of the oil before mounting, otherwise the balsam will eventually darken. The material for sections was fixed with hot (80° C.) corrosive sublimate (saturated corrosive sublimate in 35% alcohol plus 2% of glacial acetic acid) for fifteen minutes to two hours. This was replaced by 707( alcohol containing a few drops of iodine, and the material was allowed to remain in this for twenty-four or more hours. Paraffin hav- 179] HEAD OF DIPTERA — PETERSON 9 iug a melting: point of 62-6-4 C. was a sufficiently firm medium in which to cut sections as thin as eight microns. Specimens stained in toto gave the best results. Delafield's haematoxylin required 24—48 hours, and borax carmine 3-7 days. ACKNOWLEDGMENTS This investigation was carried on under the supervision of Dr. A. D. MacGillivray, and to him I am greatly indebted for the sincere interest shown and the many valuable suggestions received. Many speci- mens, unobtainable in this vicinity, were secured from the collections of the Illinois State Laboratory of Natural History, and for these I am indebted to Professor S. A. Forbes. I am indebted to the Graduate School of the University of Illinois for funds used in purchasing speci- mens. I am also indebted to Mr. J. E. Malloch, of the Illinois State Laboratory of Natural History, for the identification of all my material and for specimens and many suggestions ; to Mr. J. M. Aldrich for species of Diopsidae, Phycodromidae, and Blephai'oceridae ; to Professor A. L. Melander for a species of Cyrtidae ; to Mr. 0. S. Westcott for a species of Phycodromidae ; to Dr. P. S. Welch for a species of Simulii- dae ; and to Dr. 0. A. Johannsen for species of Dixidae and Blepha- roceridae. I am also indebted to many others wlio furnished me with unnamed material. MATERIALS Tlie following list of insects includes all of the identified forms studied. The families of Diptera to which these species belong are arranged according to Aldrich 's "Catalogue of North American Dip- tera". The generic and specific names of all but a few species may likewise be found in this catalog. Aldrich lists fifty-nine families; of these, one or more representa- tives of fifty-three families have been studied. Tlie following are not represented : Orphnephilidae, Acanthomeridae, Nemestrinidae, Apio- ceridae, Rhopalomeridae, and Nycteribiidae. The male and female of each species have been observed except in a few eases ; in these the word "male" or "female" after the species name indicates which sex has been seen. Excepting one or two forms, the male and female have both been drawn if they were decidedly different. If the two sexes are similar, the figures were mostly made from the female. An asterisk before the name of a species indicates that this form has been embedded, sectioned, and studied. The figures following the various species refer to the drawings made of the same. 10 ILLIXOIS BIOLOGICAL MONOGRAPHS [180 DiPTERA Suborder Proboscidea Orthorrhapha-Nemocera. Tipulidae.— *Tipula bieornis (Fig. 18, 95, 178, 277, 383, 384, 388, and 503), Tipula cunctans, Tipula abdominalis, Limuobia im- matura, female (Fig. 93, 386, and 507), Helobia punctipen- nis, female (Fig. 385), Trichoeera bimaeula, male (Fig. 16, 78, 158, 200, 260, 311, 365, 499, and 500), Geranomyia canadensis, male (Fig. 382 and 506), Ptychoptera rufocincta (Fig. 15), and Bittacomorpha clavipes, male (Fig. 85 and 389). Dixidae.— Dixa clavata (Fig. 19, 79, 163, 199, 262, 375, 387, 501, and 502), and Dixa modesta (Fig. 254). Psyehodidae.— Psyehoda albipennis (Fig. 8, 82, 166, 202, 263, 318, 372, 529, and 530), and Psyehoda sp. Chironomidae. — Chironomus ferugineovittatus (Fig. 12, 88, 89, 152, 206, 207, 270, 312, 371, 531, and 532), Culieoides sanguisugus (Fig. 253, 265, and 521), and Forcipomyia cilipes. Culicidae.— Psorophora ciliata (Fig. 10, 26, 96, 159, 210, 211, 251, 266, 373, 380, 381, 504, and 505), Anopheles sp., and *Culex sp. Myeetophilidae.— Sciara varians (Fig. 17, 81, 150, 205, 267, 314, 360, 512, and 513), Mycetobia divergens (Fig. 7, 90, and 161), Myeetophila punctata (Fig. 11 and 87), and Leia oblectabilis (Fig. 368). Cecidomyiidae.— Eabdophaga strobiloides (Fig. 6, 86, 170, 201, 268, 313, 367, 510, and 511), and Cecidomyia sp. Bibionidae.— Bibio femoratus (Fig. 13, 14, 91, 92, 153, 154, 208, 264, 315, 364, 522, and 523), and Bibio albipennis. iSimuliidae. — Simulium venustum, female (Fig. 2, 77, 144, 204, 250, 258, 316, 366, 489, 497, and 498), Simulium johannseni (Pig. 3 and 252), Simulium pecuarum, and Simulium jenningsi. Blepharoceridae. — Bibioeephala elegantula (Fig. 4, 5, 76, 83, 155, 156, 203, 256, 269, 399, 526, and 527), and Blepharocera sp. Ehyphidae.— Rhyphus punctatus (Fig. 9, 80, 157, 209, 261, 321, 374, 508, and 509). Orthorrhapha-Brachycera. Stratiomyiidae.— Stratiomyia apieula (Fig. 27, 28, 104, 160, 213, 273, 331, 395, 396, 545, and 546), and Stratiomyia meigeni. Tabanidae.— Tabanus giganteus (Fig. 20, 21, 74, 75, 142, 143, 214, 255, 259, 283, 317, 390-392, and 491-496), Tabanus sulcifrons, Tabanus atratus, Tabanus trimaculata, and Chrysops striatus. Leptidae.— Leptis vertebrata (Fig. 34, 35, 103, 145, 218, 275, 323, 369, 181] HEAD OF DIPTERA— PETERSON 11 370, 520, and 525), Chrysopila proxima, Chiysopila thoracica, Chrvsopila quadrata, and Chrysopila velutina. Cyrtidae.— Oncodes costatus (Fig. 53, 105, 109, 220, 486, and 487), Eulonchus tristis (Fig. 284a, 364a, 425a, 425b, and 543), and Pterodontia flavipes. Bombyliidae.— Exoprosopa fasciata (Fig. 29, 98, 162, 216, 285, 361- 426-429, 549, and 550), Systoeehus vulgaris, Lepidophora sp., and Bombylins major (Fig. 482). Therevidae. — Psilocepliala liaemorrhoidalis (Fig. 33, 36, 100, 173, 281, 324, 402, 403, 533, and 534). Seenopinidae. — Scenopinus fenestralis (Fig. 41, 42, 107, 149, 219, 282, 325, 400, 401, 537, and 538). Mydaidae.— Mydas clavatus (Fig. 30, 99, 146, 212, 271, 319, 397, 398, 535, and 536). Asilidae.— Promachus vertebra tus (Fig. 22, 84, 147, 148, 217, 276, 322, 376-379, and 517-519), Asilus notatus, and Derorayia um- brina. Dolichopodidae.— Doliehopus bifractns (Fig. 43, 112, 168, 226, 284, 432-434, 524, and 528), Doliehopus sp. (Fig. 108), Psilopodi- nus sipho, and Sympycnus lineatus. Empididae.— *Empis clausa (Fig. 26, 40, 97, 164, 215, 274, 352, 421-423, 547, and 548), Rhamphomyia glabra (Fig. 424 and 425), and Euhybus sp. Lonehopteridae.— Lonchoptera lutea (Fig. 37, 102, 177, 223, 280, 320, 406-408, 539, and 541). Phoridae.— Aphiochaeta agarici (Fig. 31, 111, 174, 224, 278, 335, 393, 394, 540, and 544), Metopina sp., and Dohrniphora con- cinna. Cyclorrhapha-Atlierieera. Platypezidae.— Platypeza velutina (Fig. 32, 110, 165, 222, 272, 326, 415, 416, 542, and 542a). Pipunculidae.— Pipunculus cingulatus (Fig. 38, 39, 106, 151, 243, 279, 327, 435, 436, 561, and 562). Syrphidae.— Eristalis tenax (Fig. 23-25, 113, 167, 232, 286, 328, 441-443, 587, and 588), Syritta pipiens, and *Allograpta ob- liqua. Conopidae. — Conops brachyrhynehus (Fig. 67, 117, 186, 221, 305, 356, 417-420, 591, and 592), Stylogaster biannulata (Fig. 359), and Physocephala tibialis. Cyclorrhapha-Calyptratae. Oestridae.— Gastrophilus equi (Fig. 54, 138, 239, and 490-492). Taehinidae.— Archytas analis (Fig. 68, 124, 197, 247, 309, 353, 468, 469, 604, and 605), Siphona geniculata (Fig. 355 and 458), 12 ILLINOIS BIOLOGICAL MONOGRAPHS [182 Gonia capitata, Oeyptera carolinae, and Gymnosoma fuliginosa. Dexiidae.— Thelaira leucozona (Fig. 65, 128, 196, 230, 301, 346, 473, 474, 595, and 596). Sarcophagidae. — Sarcophaga haemorrhoidalis (Fig. 66, 130, 191, 244, 310, 350, 477, 478, 602, and 603). Muscidae.— *Musea domestica (Fig. 71, 72, 133, 194, 242, 304, 351, 465-467, 600, and 601), Calliphora vomitoria (Fig. 484 and 485), *Stomoxys caleitrans (Fig. 354, 479, 480, and 599), Myios- pila meditabunda (Fig. 120), Pollenia rudis, Lucilia caesar, and Calliphora erytlirocephala. Anthomyiidae.— Hydrotaea dentipes (Fig. 69, 70, 127, 195, 241, 308, 349, 475, 476, 597, and 598), Lispa nasoni (Fig. 116 and 481), Dexiopsis lacteipennis, Coenosia aurifrons, and Chortophila sp. Cyclorrhapha-Aealyptratae. Seatophagidae.— Scatophaga furcata (Fig. 62, 135, 193, 246, 307, 857, 470-472, 593, and 594). Heteroneuridae.— Heteroneura flaviseta (Fig. 49, 126, 176, 229, 298, 340, 459, 460, 589, and 590). Helomyzidae.— Oecothea fenestralis (Fig. 48, 137, 192, 227, 290, 332, 452, 453, 580, and 581). Borboridae.— Borborus eqiiinus (Fig. 63, 136, 188, 231, 294, 342, 437, 438, and 565-567), Limosina ferruginata, and Sphaerocera piisilla. Phycodromidae.— Coelopa vanduzeii (Fig. 58, 121, 182, 288, 337, 448, 449, 559, and 560). Seiomyzidae.— Tetanocera plumosa (Fig. 55, 119, 180, 225, 302, 344, 463, 464, 584, and 586), and Sepedon fuscipennis. Sapromyzidae. — Sapromyza vulgaris (Fig. 60, 115, 171, 248, 289, 329, 409, 410, 553, and 554), Sapromyza bispina, Miuettia lupu- lina, and Lonchaea polita. Ortalididae.— Chrysorayza demandata (Fig. 64, 134, 181, 245, 295, 341, 456, 457, 557, and 558), Tritoxa incurva, Chaetopsis aenea, Camptoneura picta, Pyrgota sp., and Eumetopia sp. Trypetidae.— Euaresta aequalis (Fig. 61, 131, 175, 240, 292, 347, 413, 414, 572, and 573), Trypeta alba, and Straussia longipen- nis. Mieropezidae.— Calobata uuivitta (Fig. 44, 114, 183, 236, 296, 348, 446, 447, 551, and 552). Sepsidae.— Sepsis violacea (Fig. 46, 118. 184, 234, 287, 334, 439, 440, 582, and 583), and Prochyliza xantliostoma. Psilidae.— Loxoeera peetoralis (Fig. 59, 123, 169, 235, 300, 339, 461, 462, 570, and 571). 183] HEAD OF DIPTERA—PETERSOX 13 Diopsiclae.— Sphyracephala bicornis (Fig. 52. 94, 190, 293, 338, 450, 451, and 585). Ephydridae.— Ochthera mantis (Fig. 56, 101, 187, 237, 297, 336, 444, 445, 483, and 574—577), Paraliinna appendiculata, and Parydra bituberculata. Oscinidae.— Chloropisca glabra (Fig. 51, 132, 189, 306, 345, 430, 431, 555, and 556), Siphonella abdoniinalis, and Hippelates flavipes. Drosopliilidae. — Drosopliila ampelophila (Fig. 45, 125, 172, 238, 291, 343, 454,455, 563, and 564). Geomyzidae.— Chyromya concolor (Fig. 50, 122, 179, 233, 299, 333, 411, 412, 568, and 569). Agromyzidae.— Desmometopa latipes (Fig. 47, 129, 185, 228, 303, 330, 404, 405, 578, and 579). Suborder Eproboseidea Hippoboscidae.— Olfersia ardeae (Fig. 57, 139, 198, 249, 358, 488, and 606), and Melophagus ovinus. Orthoptera Periplaneta orientalis (Fig. 514). IMelanoplus differentialis (Fig. 515). Grj'llus pennsylvanicus (Fig. 516). Hypothetical and typical figures (Fig. 1, 73, 140, 141, 199h, 256h, 257, 362, 363, and 493). FIXED PARTS OF THE HEAD A hypothetical head-capsule of Diptera (Fig. 1) has a dorso-ventral extension. The epicranial suture (e. s) is present on the meson, and extends from the occipital foramen (o. f ) to a point on the cephalic aspect ventrad of the antennae. At this point it bifurcates and the two arms continue to the invaginations of the anterior arms of the tentorium (i. a), which are situated at the dorso-lateral angles of the clypeus (c). The three unpaired sclerites included within, or ventrad of, the fork of the epicranial suture are the front (fr), clypeus (c), and labnim (1). The fronto-elypeal suture is represented by a dotted line in the figure. The vertex (v) includes all of the dorsal and cephalic aspects of the epicranium except the front (fr), while the genae (ge) are the regions of the vertex ventrad and mesad of the compound ej'es. Two large compound eyes (c.e) cover the lateral portions of the cephalic aspect. Three ocelli (oc) are situated on the vertex. The occiput (occ) and postgenae (po) constitute the caudal aspect of the head-capsule. 14 ILLINOIS BIOLOGICAL MONOGRAPHS [184- The tentorium (t) of the hypothetical head-capsule has three pairs of invaginations, homologous with the invaginations in generalized in- sects. The invaginations of the posterior arms (i.p) of the tentorium are situated ventrad of the occipital foramen at tlie distal ends of chitin- ized thickenings. The invaginations of the dorsal arms of the tento- rium (i.d) are on the cephalic aspect near the antennae and adjacent to the epicranial suture, while the invaginations of the anterior arms, of the tentorium (i.a) are situated in the epicranial suture and adjacent to the dorso-lateral angles of the clypeus. The heads of all Diptera have a dorso-ventral extension, and in this respect resemble the heads of many generalized insects. Some of the primary sutures, sclerites, and invaginations of the head of such an insect are present in a number of tlie Nematocera and in a few of the Brachycera. The hypothetical head-capsule has been constructed from these forms. The heads of the Acalyptratae and the Calyptratae are highly specialized by the modification, union, reduction, and membra- nous development of parts, consequently very few if any primary char- acters remain which can be homologized with these structures. The membranous development of areas has been the most important process of specialization. The stippled areas on the figures show the extent of the membrane. The various parts of the head-capsule are discussed individually and in the order in which they were described for the hypotlietical type. The heads of Diptera naturally fall into two groups according to the presence or absence of a frontal suture (fr.s) and a ptilinum (pt). The forms without a frontal suture are the more gen- eralized. Epicranial Suture. — The epicranial suture of all insects originates in the embryo. The stem of the suture on the dorso-meson represents the line along which the paired parts of the head meet, while the arms of the suture (a. e. s) represent the place of contact between the paired sclerites of the head and the mesal unpaired sclerites. The epicranial suture (e. s) of a hypothetical dipterous head corresponds to the above description, and is homologous with the epicranial suture found in the heads of generalized immature and adult insects of the more com- mon orders. The following examples illustrate the homology between the hypothetical type and other insects. The epicranial suture in the larva of Corydalis, and in the generalized larvae of the Coleoptera, Lepi- doptera, and certain Ilymenoptera, is complete, and its two arms join with the margins of the clypeus, as in the h.ypothetical type. The epicranial suture of the adults of the Orthoptera, Hemiptera, and Hymenoptera also resembles this suture in the hypothetical head, providing the following interpretation of this suture is accepted. In 185] HEAD OF DIPTERA— PETERSON 15 the adults of Gryllus and Periplaneta it is complete and similar to that of Corydalis except that a small portion of each arm is wanting about the antennae and the lateral ocelli. The ventral ends of the arms are commonly called the fronto-genal sutures, and they join with the clypeus as in Corj-dalis. All insects that have a sucking type of mouth, such as the Hemiptera and Hymenoptera, usually show no signs of the stem of the epicranial suture. The arms, however, are distinct and form the ■lateral and dorsal boundaries of the large mesal piece commonly called the clypeus. A large number of the Diptera possess an epicranial suture which closely resembles that of the Hemiptera and the Hymenoptera. On the basis of the above interpretation of the epicranial suture it has. been possible to homologize the sutures and sclerites, and the invagina- tions of the tentorium on the cephalic aspect. No other interpretation gave satisfactory results. The epicranial suture (e. s) in Mycetophila (Fig. 11) is complete and closely resembles the hypothetical type. In Leia it closely resem- bles that of Mycetophila except for the stem of the suture, which is wanting dorsad of the median ocellus. The stem of the epicranial suture in Psorophora (Fig. 10 and 26) and Chironomus (Fig. 12) is repre- sented by a distinct suture in a deep fold on the meson. Other forms, such as Rhabdophaga (Fig. 6), Mycetobia (Fig. 7), and Tabanus (Pig. 20), show depressions or thickenings along the meson. These marks may have no significance. Outside of the above-mentioned forms, the stem of the epicranial suture is wanting. The arms of the epicranial suture (a. e. s) are present in many Diptera. This is the case in all but a few of the Nematocera, in a ma- jority of the Brachycera, and in many of the families of the Cyclorrha- pha. These resemble, therefore, the adults of the Hemiptera and Hy- menoptera. The arms are present as definite sutures between two chitinized areas in Tabanus (Fig. 20 and 21) and Leptis (Fig. 33), and in the female of Simuliura (Fig. 2). The epicranial suture is ap- parently wanting in the male of Simulium (Fig. 3) unless the lateral margins of the convex area represent it. In many genera the epicranial suture is represented by the edge of a chitinized sclerite. This is the case in Chironomus (Fig. 12), Trichocera (Fig. 16), Psorophora (Pig. 10), Mycetobia (Fig. 7), and Dixa (Pig. 19). The vertex in the genera just named is membranous between the antennal fossae and the epicra- nial suture. Sciara (Pig. 17), Rhabdophaga (Pig. 6), Bibiocephala (Fig. 4 and 5), and possibly Rhyphus (Fig. 9) and Bibio (Fig. 14), have the arms of the epicranial suture represented by the chitinized margin of the vertex, which is adjacent to the membranous portion of the fronto-clypeus. The location of the invaginations of the arms of the 16 ILLINOIS BIOLOGICAL MONOGRAPHS [186 tentorium usually helps to determine the location of the epicranial su- ture. In Ptychoptera (Fig. 15) the invaginations of the anterior arms of the tentorium are located in the distinct V-shaped depression on the chitinized area ventrad of the antennae. Undoubtedly this depression marks the position of the epicranial suture. Tipula (Fig. 18) has a very specialized head and shows no epicranial suture or tentorium. Only the arms of the epicranial sutures are present in the Brachyc- -era. On the whole these sutures are not as well developed in the Brachycera as in the Nematocera. When present (a. e. s) they are long and slit-like in all the genera except Tabanus. This condition is due to the fusion of the invaginations of the dorsal arms and the anterior arms of the tentorium along each suture. The arms of this suture in Tabanus (Fig. 20 and 21) unite the invaginations on each lateral half of the head, but they are not decidedly slit-like. The arms of the epicranial suture (a. e. s) in Tabanus (Fig. 20) have the usual inverted-u shape and their ventral ends terminate at the ventral margin of the head. The arms are indistinct ventrad of the invaginations of the anterior arms of the tentorium. The invaginations (i. a) in Promachus (Pig. 22) are slit-like and situated near the ventro- lateral angles of the compound eyes. The epicranial suture is wanting dorsad and ventrad of the invaginations of the anterior arms, and in this respect Promachus differs from Leptis and Tabanus. From Leptis (Fig. 35) it is possible to homologize the arms of the epicranial suture of all the Brachycera and those of the Cyclorrhapha. The arms of the suture in Leptis are long and slit-like and coincide with the invagina- tions of the tentorium on the cejihalic aspect of the head. They extend dorsad from the ventral margin of the head to a point ventrad of the antennae, where they unite and enclose a convex mesal area called the fronto-clypeus (fr. c). This siiture (a. e. s) in Platypeza (Fig. 32) closely resembles that of Leptis. The dorsal ends of the arms of the epicranial suture are wanting in Psilocephala (Fig. 36), Mydas (Fig. 30), Exoprosopa (Fig. 29), Eristalis (Fig. 23 and 25), and Scenopiuus (Pig. 41 and 42), and in other forms. Scenopinus shows a striking variation in that the vertex is membranous between the antennae and the fronto-clypeus, and no epicranial suture can be traced thru the membrane. Stratiomyia (Fig. 27) shows a unique development of the slits in that they extend mesad rather than dorsad. This condition is undoubtedly a secondary development. The epicranial suture of Lon- choptera, Aphiochaeta, Pipunculus, and Empis is discussed under fronto- clypeus. No epicranial suture or slit-like invaginations are present in any dipteron that has a frontal suture (fr. s) or a ptilinum (pt). Since 187] HEAD OF DIPTERA— PETERSON 17 the tentorium on the cephalic aspect and the arms of the epicranial suture are usually closely associated in insects, there is every reason to believe that the tentorial thickenings (t. th) mark the course of the suture (a. e. s). Furthermore, the location of the thickenings of the tentorium is very similar to the location of the slit-like invaginations of Leptis (Fig. 35). These thickenings (t. th) have been considered as marking the course of the arms of the epicranial suture. The extent of the tentorial thickenings varies considerably, as shown in the figures. In Tetauocera (Fig. 55), Chloropisca (Fig. 51), Heteroneura (Fig. 49), and others, the tentorial thickenings extend to the antennal fossae (a. f). No sutures are present between the dorsal ends of these thickenings. Fronto-clypeus. — The front (fr) and clj^aeus (c) of all insects are unpaired selerites located between the arms of the epicranial suture (a. e. s). The labnim (1) is also an unpaired sclerite attached typically to the ventral margin of the clypeus. These three selerites and their parts are not always distinguishable. This is particularly true of the front and clypeus in Diptera. The dotted, transverse line uniting the invaginations of the anterior arms of the tentorium (i. a) in the hypo- thetical head indicates the position of the fronto-elypeal suture. In a few of the Orthorrhapha, suture-like marks, depressions, or thickenings extend across the ehitinized portion of the fronto-clj^eus. These marks in Chironomus (Fig. 12), Mj'cetophila (Fig. 11), and Rhabdophaga (Fig. 6) resemble the fronto-elypeal siiture as indicated in the hypo- thetical type. It is possible that they are remnants of this suture. Excepting in the forms named, one can not be sure of the presence of a f ronto-ch^eal suture ; consequentlj' the entire area between tlie labrum and the arms of the epicranial suture has been designated as the fronto- clypeus (fr. c). The absence of the fronto-elypeal suture in Diptera is not unusual, since it is wanting in many generalized insects. For those who may wish to di\'ide the fronto-clj'peus into two areas, the dorsal half would be the front and the ventral half the clypeus. A large portion of the fronto-clypeus is membranous in Rhabdophaga (Fig. 6), Rhyphus (Fig. 9), and Seiara (Fig. 17), and the ehitinized part is greatly reduced. The variations found in the Nematocera are rep- resented in the figures. The Brachycera show two lines of development in the modification of the area enclosed by the arms of the epicranial suture. Both of these started from a form which possessed an epicranial suture similar to that of Leptis (Fig. 35). The line of development seen in Psilocephala, Platypeza, Scenopinus, Lonchoptera, and Aphiochaeta is considered first. The ehitinized fronto-clypeus of Leptis resembles the fronto-clypeus of a number of the Nematocera, as Seiara (Fig. 17). From tliis simple 18 ILLIKOIS BIOLOGICAL MONOGRAPHS [188 condition it is possible to develop the type of fronto-clypeus found in Psilocephala (Fig. 33 and 36). This came about by a membranous development on the meson and on the lateral margins of the fronto- clypeus and the loss of the arms of the epicranial suture directly ven- trad of the antennae. The membranous development of the fronto- clypeus of Platypeza (Fig. 32) resembles that of Psilocephala. Sceno- pinus (Fig. 41 and 42) belongs to this same line, but in this genus the antennae are adjacent to the fronto-clypeus and no portion of the ehitinized vertex exists between them. The form of the chitiuized portion of the fronto-clypeus resembles closely that of Platypeza (Fig. 32). Aphiochaeta (Fig. 31) and Lonchoptera (Fig. 37) apparently belong to this same series. If such is the case, the arms of the epici'auial suture do not project dorsad but are represented by the nearly straight ventral margin of the cephalic aspect. This condition must have come about by the straightening out of the usual u-shaped depression, and the ehitinized part of the fronto-clypeus is located ventrad of the mar- gin of the head. The tentorial thickenings along the ventral margin of the head in Lonchoptera afford evidence favorable to the above inter- pi'etation. A similar type of development occurs in Bibio (Fig. 14), in ■whicli the invaginations for the anterior arms of the tentorium are located on the ventral margin of the head-capsule latero-ventrad of the antennal fossae. All the other Brachyeera and Cyclorrhapha figured, show the presence of sclerites designated as the tormae and located ventrad of the fronto-clypeus, and this fact places them in tlie line of specialization which leads toward a muscid type. The fronto-clypeus (fr. c) is present in all Diptera and constitutes a prominent portion of the head-capsule. In Tabanus (Fig. 20 and 21) the fronto-clypeus is the entire area ventrad of the epicranial suture and outside of the tormae and the labrum. The sutures separating the fronto-clypeus from the genae (ge) are very indistinct. No arms of the epicranial suture are present in Promachus (Fig. 22), Empis (Fig. 40), and Pipunculus (Fig. 38) ; consequently the dorsal extent of the fronto-clypeus can not be determined, and the area ventrad of the antennae is considered as the fronto-clypeus. The fronto-clypeus of My das (Fig. 30) resembles that of Leptis, and from a type similar to Mydas it is possible to develop the fronto-clypeus of Exoprosopa (Fig. 29), Eristalis (Fig. 25), and probably Stratiomyia (Fig. 27). The fronto-clypeus of Mydas closely resembles that of the Acalyptratae and tlie Calyptratae, as will be seen by comparing Mydas with Tetanocera (Fig. 55), Chloropisca (Fig. 51), Chyromya (Fig. 50), and Musca (Fig. 72). It is not a completely ehitinized area in all of the genera studied, and the significance of this mesal membranous area in Sepsis, Oecothea, and Calobata has been suggested in the discussion on the ptilinum. 189] HEAD OF DIPTERA— PETERSON 19 Tonnac. — The tormae (to) in generalized insects are cliitioized pieces which belong to the lateral portions of the epipharynx in the region of the clypeo-labral suture and connect with the cljT)eus or la- brum at the lateral ends of the suture. These are well illustrated in such Orthoptera as Periplaneta (Pig. 514), Melanoplus (Fig. 515), and GryUus (Fig. 516). The tormae of generalized Diptera also connect with the inner sur- face of the ventral portion of the frouto-clypeus. They are not well- developed structures or readily distinguishable from the fronto-cl.vpeus in a number of species of the Nematocera. This seems to be due to the decidedly convex nature of the fronto-clypeus and the close proximity of its lateral portions to the lateral margins of the epipharynx. The tormae of Leptis (Fig. 520), Psilocephala (Fig. 36 and 533), Scenopi- nus (Fig. 41 and 538), Aphiochaeta (Fig. 31 and 544) Louchoptera (Fig. 37 and 539), and Platypeza (Fig. 32 and 543) connect with the fronto-clypeus and thus resemble the Nematocera and the hypothetical type. In Tabanus, the tormae (Fig. 494) resemble the above genera in their connection with the fronto-clypeus, but they have been enlarged ventrad until they are exposed between the clypeus and the labrum (Fig. 20 and 494). The exposed portions of the tormae resemble two small, triangular sclerites with their pointed ends meeting on the meson. This condition is not unusual since they resemble closely the exposed portions of the tormae located at the lateral ends of the clypeo-labral suture in Gryllus (Fig. 516). Siraulium (Fig. 2 and 489) also shows exposed portions of the tormae at the ventro-lateral angles of the fronto- clypeus (fr. c). The inverted chitinized V-shaped piece ventrad of the fronto- clypeus in Mydas (Fig. 30) has undoubtedly been derived from the fusion of the tormae of some form resembling Tabanus (Fig. 20). The tor- mae are adjacent to the fronto-clypeus in Mydas, but they are not con- nected with the same as in Tabanus. From the type of tormae found in Mydas it is possible to develop the tormae of all other genera. The tormae vary in shape and position as seen in the cephalic views of the head. In Exoprosopa (Fig. 29), Eristalis (Fig. 25), and Stratiomyia (Fig. 27) they show a striking development in that they are located within deep emarginations of the ventral nuirgin of the fronto-clypeus. The tormae of Empis (Fig. 40) closely resemble those of IMydas and belong to the same line of development. In Pipunculus (Fig. 38) the tormae resemble the fronto-clypeus of Sciara (Fig. 17), but as a matter of fact the fronto-clypeus is the area ventrad of the antennae, as shown by the location (Fig. 151) of the dorsal arms of the tentorium (d. a). The tormae of the Acal.vptratae are usually crescent-shape, while in the Calyptratae they resemble the type found in Mydas. 20 ILLIXOIS BIOLOGICAL MONOGRAPHS [190 Ptillnum. — A deep, inverted U-shaped groove is present in the heads of all the Calyptratae and the Acalyptratae dorsad of the anten- nae. This groove is called the frontal suture (fr. s) and marks the line of invagination of the large membranous pouch, the ptilinum (pt). In Sphyracephala (Fig. 52) the frontal suture is V-shaped, owing to the peculiar development of the head. The extent of the invagination of the ptilinum (pt) is indicated by a dot-and-dash line in the drawings of the cephalic and lateral views of the head-capsule. The origin of the ptilinum has been a mystery to morphologists. After a careful examination of the heads of the Brachyeera and the Cyelorrhapha, no definite data were found which would throw any light on its origin. A few forms, however, suggested a possible way in which it might have been developed. The frontal suture and the ptilinum are comparatively small in Tetanocera (Fig. 55), Sapromyza (Fig. 60), Conops (Fig. 67), Oehthera (Fig. 56), and Chloropisea (Fig. 51). These genera gave no clue to the early stages of its development unless the thinly chitinized condition of the fronto-cl.ypeus of Chloropisea has some significance. It seems evident that the frontal suture was once a membranous area which became invaginated to form a membranous pouch or ptilinum. If this is the case, the mesal membranous area of the fronto-clypeus of Sepsis (Fig. 46), Oecothea (Fig. 48), Calobata (Fig. 44), and Desmometopa (Fig. 47) would be very significant. The ptilinum might possibly have originated from some form similar to Scenopinus (Fig. 41), in which the ventral margin of the chitinized vertex is located dorsad and laterad of the antennae. It seems quite possible that the membrane along this margin became invaginated in the early stages of the development of the ptilinum. The above con- jectures may or may not be correct. A real solution of the problem will undoubtedly require a careful study of the pupal development. Lahrum. — The labrum (1) of a hypothetical dipterous head (Fig. 1, 140, and 493) is a distinct, chitinized, tongue-like structure connected with the ventral margin of the clypeus. The shape and size of the labrum are identical with the shape and size of the epipharynx, which is located on its caudal aspect. The labrum {1) and epipharynx (ep) are joined together by a membrane along their lateral margins. These two structures thus act as one organ and they have rightly been called the labrum-epipharynx (1. ep). The above relation of the labrum to the epipharynx and the fronto-clj^jeus resembles that in the Orthoptera. In a general way the labrum of all the genera studied resembles the hypothetical type described above. It varies, however, in shape and in degree of ehitinization. In Promachus (Fig. 22), in Psorophora (Fig. 10 and 26), and in the female of Tabanus (Fig. 20) it is completely 191] HEAD OF DIPTERA— PETERSON 21 eliitinized aud separated from the fronto-elypeus by a suture. In all other genera there is a distinct membranous area present between the fronto-elypeus and the labrum. This area is very extensive in the Cyclorrhapha and includes the ectal exposure of the tormae. The la- brum of a few scattered genera, such as Rhabdophaga (Fig. 6), Myeeto- bia (Fig. 7), Chironomus (Fig. 12), Seenopinus (Fig. 41), and others, is completely membranous, while in still others it is nearly so, as in Mydas (Fig. 30). The figures of the cephalic aspect of the head and the lateral views of the epipharynx aud the hypopharynx show the shape and extent of the chitinization of the labrum. The labrum of Dixa (Fig. 501), Trichocera (Fig. 499), Sciara (Fig. 513), Bibio (Fig. 523), Simulium (Fig. 497), Culicoides (Fig. 521), Tabanus (Fig. 20), and Dolichopus (Fig. 528) is di.stinctly sepa- rated from the epipharynx (ep) by a membrane. This condition is best seen in a lateral view. A majority of the forms studied have little or no membrane between the labrum and epipharynx. This is particu- larly true of the Cyclorrhapha. The surface of the labrum of all Dip- tera is more or less convex. In a large number of the genera the con- vexity is very decided and of such a nature as to surround the cephalic and lateral aspects of the epipharynx. The epipharynx in these forms can only be seen in a caudal view. In the Calyptratae, the labrum and epipharynx are firmly united in one piece. The labrum of Simulium (Fig. 2 and 489) is unique in that the chitinized part consists of a narrow mesal piece which bifurcates at its distal end. These bifurcations give rise to special small liook-like struc- tures (h) which have been incorrectly interpreted as mandibles (Smith, 1890). The labrum and epipharynx of Psorophora (Fig. 504) fit to- gether very closely. By careful dissection they may be separated, as seen in the drawing. So far as observed, no membrane is present be- tween them. The proximal end of the labrum is crook-like in form, and muscles connect with this portion. Vertex. — The vertex (v) of a hypothetical head (Fig. 1) consists of the paired continuous areas on the cephalic aspect of the epicranium. It is interpreted as including all the cephalic and dorsal aspects of the epicranium except the front. In a number of the Diptera, as heretofore described, the stem of the epicranial suture (s. e. s) is present and marks the line of fusion of the two halves of the vertex, upon which the ocelli and the antennae are located. The shape and size of the chitinized portion of the vertex is largely determined by the size of the compound eyes, the location and extent of the membranous area about the base of the antennae, and the location of the arms of tlie epicranial suture. The variations in the size and shape of the vertex are .shown in the figures of the cephalic aspect of the head. 22 ILLINOIS BIOLOGICAL MONOGRAPHS [192 The region of the vertex ventrad and mesad of each compound eye is a gena. The size of the genae (ge) is dependent upon the location of the compound eyes and the ventral extension of the head-capsule. The figures show considerable variation in these respects. Compound Eyes and Ocelli. — The compound eyes (c. e) of a hypo- thetical head are large oval structures located on the cephalo-lateral aspects of the head-capsule. They cover from one-half to two-thirds of the entire cephalic aspect and their caudal margins are adjacent to the lateral margins of the head. The compound eyes of a majority of the Diptera resemble in general the hypothetical type. The shape and size vary considerably with the different species. Variations are most prevalent in the families of the Orthorrhapha. This variability agrees with the decided variability of other parts. In such genera as Tipula (Pig. 95), Psorophora (Fig. 96), and Limnobia (Fig. 93) the compound eyes are exceptional in that they extend onto the caudal aspect of the liead. The variations in shape are well illustrated by the numerous figures. The compound eyes show secondary characters in a greater number of species than any other fixed or movable part. This sexual variation is most prevalent among the Nematoeera and the Brachycera, and was not observed in the Acalyptratae. Among the Calyptratae, slight differences occur in Musca (Fig. 71 and 72) and Hydrotaea (Fig. 69 and 70). When sexual variation occurs, the eyes of the male are larger than those of the female, and they are usually adjacent along a portion of their mesal margins. Such species are said to be holoptic ; while all the females, and some of the males, having the eyes distinctly separated, are dichoptic. The extent of the holoptic condition depends upon the size of the eyes and the location of the antennal fossae, as in Simulium (Fig. 2 and 3) and Bibio (Fig. 13 and 14). In the male of Bibio the compound eyes are adjacent along their mesal margin and the antennal fossae (a. f ) are located ventrad of the eyes. The extent and nature of the sexual variation is shown in the figures. Except in the case of Empis the heads of the male and female have both been drawn when decided differences are present. The facets or ommatidia of the compound eyes vary in number, form, and size thruout the order. In the Nematoeera they are usually large and not as closely compacted as in the Cyclorrhapha. An inter- esting variation occurs in the male of Simulium, the facets (fa) of the ventral half of the eye being smaller than those of the dorsal half. This difference is also foi;nd in the female of Bibiocephala (Fig. 5). In the male of Bibio (Fig. 154) the facets (fa) in the ventro-caudal portions of the eyes are smaller than the others. The compound eyes of Bibio- cephala and Blepharocera are divided into a dorsal and a ventral por- 193] HEAD OF DIPTERA — PETERSON 23 tion by a transverse constriction (eh), where the ommatidia are wanting. This constriction is also present in Bibio, but in this form it is confined to the caudo-ventral portion of the eye. The drawings of the lateral aspects of some heads show a line of dashes or a solid line around the margins of the compound eyes. This line indicates the extent of the infolding of the head-capsule adjacent to the compound eye. This infolding, or ocular sclerite (o. s), is figured only for those species in which it is closely related to the external mark- ings found on the caudal aspect dorsad of the occipital foramen. The influence of this invaginated edge will be more fully discussed later. The three ocelli (oc) of the hypothetical head-capsule (Fig. 1) are arranged in the form of a triangle and located on the cephalo-dorsal aspect of the vertex. The median ocellus is in the epicranial suture, somewhat ventrad of the lateral ocelli. In Leia it is in this suture somewhat dorsad of the bifurcation, and the other two ocelli are some- what laterad of it. This location of the ocelli in the Diptera agrees with Comstock's idea concerning the caudal migration of the ocelli in specialized insects. In generalized insects all three ocelli may be on the front or two on the vertex while the median ocellus is on the front. The ocelli in the Hymenoptera and Hemiptera are similar in location to those of the Diptera. Leia is the only form studied which has ocelli and a well-marked stem of the epicranial suture. The chitinized, secondary, Y-sliaped thickenings on the ocellar triangle of Rhyphus (Fig. 9) and Mycetobia (Fig. 7) should not be confused with the epicranial suture. Three ocelli are present in all other genera of Diptera examined except Oncodes (Fig. 53) and Mycetophila, in which there are only two. The median ocellus is wanting in Mycetophila, while the lateral ocelli are small inconspicuous bodies, adjacent to the dorso-mesal margin of the compound eyes (not shown in the figure). The figures show such variations as occur in the various ocellar groups. Occiput and Postgenae. — No sutures occur on the caudal aspect of the hypothetical head-capsule (Fig. 73) except the epicranial suture (e. s). This absence of sutures makes it impossible to locate definitely the boundaries of the occiput and the postgenae. The following in- terpretation is based iipon a study of the occiput and postgenae of generalized insects, .such as the Orthoptora. The occiput comprises aU the area dorsad of an imaginary transverse line drawn thru the middle of the centrally located occipital foramen. The areas ventrad of this line and laterad of the mesal membranous areas are the postgenae. The occiput (occ) undergoes a secondary development about the margin of 24 ILLIXOIS BIOLOGICAL MOSOGRAPHS [194 the occipital foramen. The structures pertaining to this modification have been designated as the parocciput (pocc). Each postgena (po) is also secondarily differentiated along its mesal margin by a chitinized thickening which extends between the occipital foramen and the invagi- nations of the posterior arms of the tentorium. This thickening has been designated as the parapostgenal thickening, while the area mesad of it is the parapostgena (ppo). The two mesal projections of the parocciput on the lateral margin of the occipital foramen serve as points for the articulation of neck sclerites and mark the ventral boundary of the occiput. The occipital foramen (o. f ) is centrally situated in all but a few genera, such as Tipula (Fig. 95), Limnobia (Fig. 93), Psorophora (Fig. 96), and Bibio (Fig. 92), in which it is near the dorsal margin. The size of the occipital foramen is more or less constant thruout the order, but in Psj'choda (Fig. 82) and Promachus (Fig. 84) it is comparatively much larger than in PipunciUus (Fig. 106) and Exoprosopa (Fig. 98). The shape of the occipital foramen varies somewhat, but usually it is in the form of a figure eight. The constrictions in the lateral margins are generally due to the mesal projections of the parocciput, which vary to some extent in their situation. The projections in Exoprosopa (Fig. 98), Pipunculus (Fig. 106), and Mydas (Fig. 99) meet on the meson and completely divide the occipital foramen into two openings. The neck sclerites (n. s) always articulate with these mesal projections and are represented in a number of the figures. The occiput (occ) of all genera figured resembles in general the occiput of the hypothetical head, since no sutures separate the vertex, the occiput, and the postgenae. The position of the occipital foramen and the contour of the caudal surface determine the amount of variation in the occiput as well as in the postgenae. In some genera, Empis (Fig. 164) and Bibiocephala (Fig. 156), the caudal aspect is convex; while in others, Exoprosopa (Fig. 98) and Pipunculus (Fig. 106), it is de- cidedly concave. Siiture-like markings or depressions are present near the dorsal margin of the caudal aspect in the heads of Tabanus (Fig. 74), Stratiomyia (Fig. 104), Bibio (Fig. 91), Bibiocephala (Fig. 83), Leptis (Fig. 103), Psilocephala (Fig. 100), and others. These depres- sions mark the place of contact of the mesal portions of the ocular sclerites with the head-capsule, and are in no way homologous with the sutures about the occiput in generalized insects. The area about the dorsal and lateral margin of the occipital fora- men, the parocciput (pocc), is more or less differentiated from the re- mainder of the occiput in all the species studied. In the more generalized forms, Bibiocephala (Fig. S3), Trichocera (Fig. 78), Tipula (Fig. 95)^ 195] HEAD OF DIPTERA — PETERSOX 25 Sciara (Fig. 81), and Bittacomorpha (Fig. 85), it is only a thickened edge ; but in a large nximber of species thruout the order it is a clearly defined piece, set off from the occiput proper bj- a secondary suture. The indefiniteness of this piece in a large number of the generalized Diptera and the want of an homologous part in generalized insects support the view that it is only a secondary modification of the occiput. The parocciput (poce), in most genera, occurs as a narrow piece about the dorsal and lateral margin of the occipital foramen, and its ventral ends project mesad. In the heads of the Cyclorrhapha thi-ee secondarily developed, chitinized thickenings (th) arise from the ental surface of the parocciput; two of these project dorso-laterad from the lateral portions of the parocciput, and the third is on the meson. These thickenings are also present in some of the Brachj-cera, such as Dolieho- pus (Fig. 112). Their greatest development is found in Eristalis (Fig. 113), where two dorso-lateral thickenings (th) extend to the caudal margins of the compound eyes and a third thickening, on the meson, bifurcates a short distance dorsad of the occipital foramen, the two arms connecting with the dorso-mesal angles of the compound eyes. In the genera figured, the dorso-lateral thickenings are, on the whole, better developed than the thickening on the meson. In Thelaira (Fig. 128) and ilusca (Fig. 133) the dorso-lateral thickenings project dorsad to the margin of the head. The area included between them is called by several writers the epicephalon, or the occiput; and tho it is entirely different in origin from similarly situated areas in Tabanus (Fig. 74) and other genera, the same name is applied in the different cases. These names and others used by systematists have no morphological signifi- cance for they can not be homologized with the primary selerites of a generalized insect. The postgenae (po) of the hypothetical dipterous head have been carefully compared with those of the heads of such generalized insects as the Orthoptera. The mesal membranous area between the postgenae is homologous with the membrane of the neck and with the membrane surrounding the proximal ends of the maxillae and the labium. There are no sutures or selerites along the mesal portions of the postgenae ia such generalized insects as the Orthoptera ; consequently the parapost- genae (ppo) described above can not be homologous with any primary sclerite. In Diptera the parapostgenae are undoubtedly special modi- fications of the postgenae. The postgenae and the parapostgenae of a majority of the Nematoc- era resemble those of the hypothetical head. In Chironomus (Fig. 88) and Trichocera (Fig. 78) the parapostgenal thickenings are want- ing. The invaginations for the posterior arms of the tentorium in 26 ILLINOIS BIOLOGICAL MONOGRAPHS [196 Simulium (Fig. 77) are adjacent to the occipital foramen, consequently the parapostgenae are confined to the lateral margins of the occipital foramen. In Tabanus also the invaginations are adjacent to the occipi- tal foramen, and the postgenae are connected ventrad of the occipital foramen in the male and by a narrow strip in the female. The area ventrad of the occipital foramen is a continuous ehitinized piece in all of the Cyclorrhapha and the Orthorrhapha. There is only one probable explanation of the origin of this area. It has been derived from the fusion of the mesal margins of the postgenae. The evidence for this interpretation is found in a number of the Nematocera. The mesal margins of the postgenae in Trichocera (Fig. 78) and Seiara (Fig. 81) are curved mesad and in some cases actually join, as in the female of Bibiocephala (Fig. 83). The peculiar elongated heads of Limnobia (Fig. 93), Tipula (Fig. 95), and Psorophora (Fig. 96) show a distinct depressed line on the meson along wliich the postgenae have joined. In a number of the genera of the Orthorrhapha and the Cy- clorrhapha the ventral margin of the caudal aspect is decidedly concave. This condition may be due to a former stage in the development of the fused postgenae. In all cases where the area ventrad of the occipital foramen is ehitinized, the invaginations of the posterior arms of the tentorium are somewhat adjacent to the occipital foramen and the attachments of the maxillae are removed to or beyond the ventral mar- gin of the head. Seiara (Fig. 81) is a good example of an early stage in the development of the above relationship. The variations in the shape and extent of the postgenae and the parapostgenae are well illus- trated by the figures. Tentorium. — There is present within the head of generalized insects a definite arrangement of ehitinized rods and plate-like structures which go to support the internal organs and furnish places for the attachment of muscles. These rods or plates arise from three pairs of openings on the head known as the invaginations of the anterior arms, dorsal arms, and posterior arms of the tentorium. The invaginations of the anterior arms are usually associated with the lateral margins of the clypeus, with one of the points of articulation of the mandibles, and frequently with the ventral ends of the arms of the epicranial suture. The invagi- nations of the dorsal arms are associated with the points of attachment of the antennae and near the dorsal portions of the arms of the epi- cranial suture. The invaginations of the posterior arms are associated with the occipital foramen and the points of attachment of the maxillae. The three pairs of arms i:nite within the head ; the small dorsal arms unite with the larger anterior arms, and these, in turn, join with the posterior arms, which are confined to the caudal portion of the head- 197] HEAD OF DIPTERA—PETERSOX 27 capsule. The free euds of the posterior arms are fused aud form the body of the tentorium. The tentorium undergoes a considerable amount of variation in the different orders, but so far as observed the above associations be- tween the invaginations and the fixed and movable parts of the head are always retained by the more generalized members of each order. This is also true for a generalized hpothetieal dipterous head. The tentorium (t) of such a head (Fig. 140 and 141) is considerably modi- fied when compared with the tentorium of a generalized insect. Two pairs of invaginations are present on the cephalic aspect of the head (Fig. 1). The dorsal, indistinct pair (i. d), just ventrad of the anten- nae, are homologous with the invaginations of the dorsal arms of the tentorium, while the prominent pair (i. a) of invaginations ventrad of these and located in the arms of the epicranial suture (a. e. s) and adjacent to the lateral ends of the fronto-clypeal suture are the invagi- nations of the anterior arms of the tentorium. One pair of invagina- tions (i. p) is present on the caudal aspect of the head-capsule (Fig. 73) somewhat ventrad of the ventro-lateral margins of the occipital foramen. These are the invaginations of the posterior arms of the tentorium. Each lateral half of the tentorium is Y-shaped (Fig. 141), the stem of the Y arising from the invaginations on the caudal aspect, its caudal portion being a part of the posterior arms (p. a) of the tento- rium. The large ventral arm of the Y and the cephalic portion of its stem, constitute the anterior arm (a. a), and the small dorsal arm of the Y is the dorsal arm (d. a) of the tentorium. These two arms con- nect with their respective invaginations on the cephalic aspect. The body of the tentorium (b. t) is apparently represented by a small, rudi- mentary, mesal projection arising from the posterior arms near the caudal portion of the stem of the Y. The association between the movable appendages and the invagi- nations of the tentorium is discussed under the respective appendages. From this point, the tentorial structures as they occur in the various genera are compared with the hypothetical type and the line of speciali- zation noted. The forms without a ptilinum are considered first. The parts of the free tentorium, not completely fiised with the head-capsule, are indicated in the figures by dotted lines. The tentorium of Tabanus (Fig. 142 and 143) is generalized and closely resembles the hj-pothetical type ; consequently it furiiislies a good starting point for a discussion. Two pairs of invaginations are present on the cephalic aspect (Fig. 20) ; of these the invaginations for the anterior arms (i. a) are the more prominent. The dorsal arms (i. d) arise from the head-capsule just ventro-laterad of the antennae 28 ILLINOIS BIOLOGICAL MONOGRAPHS [198 and connect with the arms of the epicranial suture (a. e. s). The in- vaginations of the anterior arms are situated near the ventral ends of the arms of the epicranial suture. The invaginations on each lateral half of the head are joined together by the arms of the epicranial suture and resemble the hypothetical type. Two pairs of invaginations are also present on the cephalic aspect of Simulium (Fig. 2 and 3), but in this genus they are not as prominent as in Tabanus. They are situated on the vertex (v), adjacent to the compound eyes. In the female the arms of the epicranial suture are well defined and the invaginations are closely adjacent to them, while in the male the sutures are wanting. Tabanus and Simulium are the only forms figured which show two distinct pairs of invaginations on the cephalic aspect. All other genera have only one pair and these are of two types. They are either long and slit-like or they resemble small pits or darkened spots on the ectal surface. The long slit-like invaginations found in Leptis (Fig. 35), Psilocephala (Fig. 36), Platypeza (Fig. 32), Seenopiniis (Fig. 41), Exoprosopa (Fig. 29), Stratiomyia (Fig. 27), Mydas (Fig. 30), Erista- lis (Fig. 25), and other genera have a special significance which will be more fully discussed later. The small, pit-like invaginations are present in the Nematocera and in Pipunculus (Pig. 38) and Empis (Fig. 40). These are situated on the chitinized area of the vertex; or on the fronto-clypeus, adjacent to the arms of the epicranial suture and usually close to the compound eyes. Their position and structure indi- cate that they are the invaginations of the anterior arms of the tento- rium. In a few of the genera of the Orthorrhapha and in some others, as Lonchoptera (Fig. 37), Tipula (Fig. 18), and Aphiochaeta (Pig. 31), no invaginations are present on the cephalic aspect of the head. One pair of invaginations, that for the posterior arms (i. p) of the tentorium, is present on the caudal aspect of the heads of all genera examined except Oncodes (Pig. 105), Olfersia (Fig. 139), Tipula (Pig. 95), and perhaps a few species of other genera in which it is difScult to be sure of their presence. These invaginations in Bibioeephala (Fig. 83), Trichocera (Fig. 76), Dixa (Fig. 79), Ehyphus (Pig. 80), Sciara (Fig. 81), Psychoda (Fig. 82), Rhabdophaga (Fig. 86), Chironomus (Fig. 88), Bittacomorpha (Fig. 85), Mycetophila (Fig. 87), and Myce- tobia (Fig. 90) are decidedly ventrad of the occipital foramen and adjacent to the proximal ends of the maxillae. They are connected with the lateral margins of the occipital foramen by means of the para- postgenal thickenings except in Chironomus and Trichocera. The above- named forms closely resemble the hypothetical type. In a few genera of the Nematocera, such as Psorophora (Fig. 96) and Simulium (Fig. 77), the invaginations are adjacent to the occipital foramen. This 199] HEAD OF DIPTERA— PETERSON 29 position is characteristic of these invaginations in the Brachycera, and the figures show the details of the variations in the position of the invaginations on the posterior arms of the tentorium. Two lines of specialization appear in the tentorium of the Diptera, one in the reduction of the dorsal arms and the other in the union of the dorsal arms with the anterior arms. The two tj-pes of invaginations described for the cephalic aspect of the head bear directly upon this problem. The most important evidence in proof of these two types of development is found in the structure of the arms. In Sciara (Fig. 150), Bibio (Fig. 153 and 154), Psorophora (Fig. 159), Trichocera (Fig. 158), Bibiocephala (Fig. 155), Dixa (Fig. 163), and others, two long narrow rods extend on each side between the invaginations on the caudal aspect and the invaginations on the cephalic aspect. These rods are composed of the posterior arms (p. a) and the anterior arms (a. a) of the tentorium. The dorsal arms are completely reduced in these forms. Other genera show completely developed dorsal arms or rudiments of the same. The dorsal arms (d. a) are distinct and free in Pipunculus (Fig. 151). They arise from the anterior arms and project eephalad to the cephalic aspect of the head, where thej'- connect with small but distinct ental projections adjacent to the anten- nae. The cephalic ends of the dorsal arms are very delicate and easily broken in dissecting. There are no invaginations on the ectal surface. In Chironomus (Fig. 152) the tentorial arms are swollen near the mid- dle of their length, and the distinct humps on the dorsal side are interpreted as rudiments of the dorsal arms. Promachus (Fig. 147) has two long, free, finger-like projections, arising from the ocular sclerite near the antennae, which project toward the tentorium proper. These projections are apparently dorsal arms of the tentorium, or derivatives of the same that have retained their connection with the ocular sclerite near the mesal margin of the compound eye but have lost their connection with the tentorium proper. A similar relationship exists between the dorsal arms and the ocular sclerite in Tabanus (Fig. 22). If the above structures in Promachus are dorsal arms, then the anterior arms are large (Fig. 148) and the slit-like invaginations on the cephalic aspect are only the invaginations of the anterior arms of the tentorium. The tentoria of the Nematocera above described are in the ventral half of the head-cavity and their situation is dependent upon the posi- tion of the invaginations. Usually the invaginations of the anterior arms are ventrad of the invaginations of the posterior arms ; but Bibio- cephala (Fig. 155) is an exception to this rule if the tentorium in this genus is composed of only the anterior and posterior arms — and there 30 ILLINOIS BIOLOGICAL MONOGRAPHS [200 is no evidence to the contrary. In some genera, as in Lonchoptera (Fig. 177), Rhabdopliaga (Pig. 170), and Empis (Fig. 164), the tento- ria are not free rods extending thru the head cavity, but are completely united with the ventral margin of the head, or nearly so. The tentorium of Aphiochaeta (Fig. 174) is reduced to two small ental projections adjacent to the occipital foramen, while in Tipula (Fig. 178) the ten- torium is apparently wanting. In a majority of the Brachycera the tentorial arms are specialized by fusion, and Tabanus (Fig. 143) illustrates an early stage in this development. The principal difference between the tentorium of Taba- nus and the hypothetical type is the presence of a thin chitinized plate in the V-shaped opening between the anterior and dorsal arms. Simu- lium (Fig. 144), of the Nematocera, has a similar plate, and these two genera clearly demonstrate the first stage in the fusion of these two arms. The cephalic end of the tentorium in My das (Fig. 146), Leptis (Fig. 145), Scenopinus (Fig. 149), and Exoprosopa (Fig. 162) is a broad uniformly chitinized triangular area. This condition is accounted for on the basis of the union of the anterior and dorsal arms. The invaginations on the cephalic aspect of these forms agree in all respects with this interpretation. In Tabanus (Fig. 20) the invaginations on each side are joined together by the epicranial siiture, while in the above forms the invaginations are slit-like and occupy the greater part of the arms of the epicranial suture. The slit-like invaginations are easily explained if the anterior and dorsal arms are considered as united. The posterior arms of the tentoria of the Nematocera and the Brachycera vary in shape, size, and location. The anterior and posterior arms are united within the head and no sharp line can be drawn be- tween them. The body of the tentorium (b. t) is represented by small projections on the mesal surface of the posterior arms of most genera. Many interesting features occur in the modifications of the tentoria of this group. In Dolichopus (Fig. 43 and 168) it appears to be fused with the dorsal margin of the slit-like openings on each side between the mesal margin of the compound eye and the fronto-clypeus. The tentorium of Mydas (Fig. 146) is large and tubular, and it is possible to push a good-sized needle thru the opening on the cephalic aspect to the opening of the posterior arms on the caudal aspect. The tentoria of the genera possessing a ptilinum differ principally from the foregoing in the degree of fusion with the head-capsule. In most genera of this group the tentorium is completely united with the head, but in a number of the Aealyptratae the tentorial arms arise as free rods from the invaginations on the caudal aspect and project to the latero-ventral margins of the head-capsule, with which they unite 201] HEAD OF DIPTERA — PETERSON 31 and continue cephalad as thickenings adjacent to the ventral margin of the head, as in Sapromj'za (Fig. 171), Loxoeera (Fig. 169), Euaresta (Fig. 175), Calobata (Fig. 183), Chrysomyza (Fig. 181), Drosophila (Fig. 172), Chyromya (Fig. 179), Heteroncura (Pig. 176), and Teta- nocera (Fig. 180). In those forms where the tentorium is completely fused with the head, as in Sepsis (Fig. 184), Chloropisca (Fig. 189), Coelopa (Fig. 182), and Borborus (Fig. 188), it is a continuous thick- ening from the latero-ventral angle of the occipital foramen to the cephalo-ventral aspect of the head-capsule. The tentorium between the invaginations for the posterior arms and the ventro-lateral margins of the head-eapsule is apparently wanting in Musca (Fig. 194), Thelaira (Fig. 196), Arehytas (Fig. 197), and some other genera; in one or two cases it is possible to trace a faint mark which would indicate the line of connection. The tentoria of some of the genera of the Acalyp- tratae and the Calyptratae show an iinusual development of the tento- rial thickenings (t. th) in that they extend about the entire caudal part of the ventral margin of the head. In some cases these tentorial thick- enings reach the occipital foramen, as in Calobata (Fig. 114), Scatoph- aga (Fig. 135), Heteroneura (Fig. 126), Lispa (Fig. 116), and Mj-ios- pila (Fig. 120), while in Musca (Fig. 133), Coelopa (Fig. 121), Hydrotaea (Fig. 127), and other genera, there is no such connection. The invaginations of the posterior arms of the tentorium of the Acalj-ptratae and the Calyptratae are situated laterad or latero-ventrad of, and adjacent to, the occipital foramen. In many of the species figured the invaginations are merely long, heavily chitinized furrows extending latero-ventrad from the occipital foramen, and very often it is difficult to locate them definitely. Two mesal projections arise from the proximal portions of the posterior arms in a majority of the Cyelorrhapha. In some species these structures are well developed, and their mesal ends apparently join on the meson, cephalad of the occipital foramen. These structures are similar to those described for the Brachycera and are rudiments of the body of the tentorium. No invaginations of the tentorium occur on the cephalic aspect in any of the forms which possess a ptilinum. On account of the decided specialization of this aspect, it is very difficult to know just what has happened. The tentorium is represented by thickenings which extend from the ventral to the cephalic aspect of the head. The extent of these thickenings varies; in some genera they continue to the antennal fossae, while in others they are practically wanting. 32 ILLINOIS BIOLOGICAL MOXOGRAPHS [202 MOVABLE PARTS OF THE HEAD In arrangement and structure the movable parts of the head of the generalized Diptera are homologous with the movable appendages of other generalized insects. In the Cyelorrhapha the parts retain their relative position, but structurally they undergo striking modifications and in some cases almost complete reduction. To make clear the use of a number of terms found in the following discussions, the mouth-parts as a whole will be considered at this point. The appendages of the mouth of the generalized Diptera are free, inde- pendent structures, with their proximal ends adjacent to the head-cap- sule. The cardines and stipites of the maxillae ai-e exceptions to the above statement, in that they are embedtled in the mesal membranous area of the caudal aspect of the head. The mouth-parts, the labrum- epipharynx, and the hypopharynx constitute in the Calyptratae a single complex mouth-appendage designated as the proboscis. The chitinized parts of the jDroboscis are far removed from the head-capsule, but in this projection of the parts, the proximal ends of the chitinized ap- pendages are joined together and have the same relationship with each other as in generalized insects. The term proboscis is most applicable among the Cyelorrhapha to those whose mouth-parts resemble those of Musca. The proboscis is naturally divided into three areas by the two bends which it makes as it is withdrawn into the oral cavity. Tlie parts of the proboscis have beeu given varied and confusing names. Hewitt divides it into two general areas — the rostrum and the proboscis proper. He says : ' ' The proboscis consists of two parts, a proximal membranous conical por- tion, the rostrum, and a distal half, the proboscis proper, which bears the oral lobes. The term haustellum is also used for this distal half (minus the oral lobes) and as a name it is probably more convenient, as the term proboscis is used for the whole structure, — rostrum, haustel- lum and oral lobes". The terms rostrum and haustellum have been used in various ways by numerous workers in diilerent orders; consequently the parts which they designate are by no means homologous. A more comprehensive set of terms based upon the word proboscis has been used by a few workers, who divide the proboscis into basiproboscis, mediproboscis, and 203] HEAD OF DIPTERA— PETERSON 33 distiproboscis. These terms have here been adopted. The basiproboscis (bpr) is equivalent to the rostrum, and may be defined as the mem- branous, cone-shaped area between the ventral margin of the head- capsule and the proximal end of the theca. The tormae, labrum- epipharynx, hypopharynx, and maxillae are parts of the basiproboscis. The mediproboscis (mpr) is the median section of the proboscis and includes the theca and the ehitinized cephalic groove of the labium. It is equivalent to the haustellum of most authors. The distiproboscis (dpr), the enlarged dilated lobes at the distal end of the proboscis, is composed of the paraglossae, with their pseudotracheal areas, and the glossae. The distiproboscis is equivalent to the oral lobes, or labellae. The movable appendages of the head are discussed in the following order: antennae, mandibles, maxillae, and labium. Antennae. — The antenna of a generalized hypothetical dipterous head (Fig. 199h) is many-segmented and of a filiform type. All the segments are similar excepting the two large proximal ones known as the scape (sc) and the pedicel (pd). The scape articulates with the ehitinized antennal sclerite (a. s) which bounds the periphery of the antennal fossa (a. f ) that is situated on the vertex dorsad of the arms of the epicranial suture. The antennae of the hypothetical type resem- ble the antennae of many generalized insects. The antennae of a majority of the Nematocera resemble the hypo- thetical type, and on the whole resemble each other. The variations in shape and size can be seen in the figures. Seeoudai'y sexual variation occurs in a few of the Nematocera, in which the antennae of the male, illustrated by Chironomus (Fig. 207) and Psorophora (Fig. 211), bear long flexible setae while those of the female are almost bare. The antennae of the Brachycera show a wide range of development, but in a majority of the genera figured the main line of specialization is toward the type found in Lonchoptera (Fig. 223) and Dolichopus (Fig. 226). One of the striking exceptions to this general line of de- velopment occurs in the geniculate type found in Stratiomyia (Fig. 213). The antennae of the Brachycera have, as a ride, fewer segments than the Nematocera. The scape and pedicel undergo only a slight change, in this group, but the flageUum (fl) is greatly modified. The proximal segment of the flagellum, or the third segment of the antenna, is enlarged, while the remaining segments are so reduced in size as to resemble the lash of a whip. The lash-like portion of the antenna is called the arista (ar). The following genera suggest the various stages thru which the antennae have passed in attaining the museid type of development. In Tabanus (Fig. 214), Empis (Fig. 215), Exoprosopa (Fig. 216), Promachus (Fig. 217), and Leptis (Fig. 218) the flagellum 34 ILLINOIS BIOLOGICAL MONOGRAPHS [204 is stylate, and the third segment is large and conical, with one or more segments at its distal end. The antennae of Platypeza (Fig. 222), Lonchoptera (Fig. 223), Aphiochaeta (Fig. 224), Oecothea (Fig. 227), and Dolichopns (Fig. 226) show an advanced stage of development in which the third segment is large and round and the remaining segments are lash-like and situated toward one side of the third segment. All but a few of the antennae of the Cyclorrhapha have apparently devel- oped from a type similar to the last-mentioned genera. The principal differences between the antennae of this group are in the length and breadth of the third segment and in the modification of the arista. The antennae of Olfersia (Fig. 249) are of a reduced museid type, and are inserted in deep cavities on the cephalic asjject of the head ; the scape and pedicel are greatly reduced, and the arista is merely a small pro- jection on the lateral aspect of the large segment. Antennal sclerites (a. s) are present only in Chironomus (Fig. 12 and 206) and Psorophora (Fig. 10 and 26). In these genera it is a distinct chitinized ring about the proximal end of the scape. The extent and place of the membrane with which the antennae are connected vary considerably. In Trichocera (Fig. 16), Chironomus (Fig. 12), Psorophora (Fig. 26), Mycetobia (Fig. 7), and some other genera it is very extensive. A general survey of the antennae of the Diptera shows that in the Nematocera they are generalized and on the whole resemble each other. The specialized antennae of the Cyclorrhapha in all but a very few genera are of a museid typ^, and also quite similar in form. The antennae of the Brachycera present a few specialized types, but the majority of them show intermediate stages between the forms found in the Nematocera and those of the Cyclorrhapha. Mandibles. — Only a few of the generalized Diptera possess mandi- bles. They are present in the females of Simulium (Fig. 2 and 250), Tabanus (Fig. 255 and 317), Psorophora (Fig. 159 and 251), Cidicoides (Fig. 253), Dixa (Fig. 254), and Bibiocephala (Fig. 155 and 256), but wanting in the males of all the species examined except Simulium (Fig. 3 and 252). The males of Simulium johannseni and S. jenningsi have distinct mandibles. No other males of Simulium were examined. So far as known this is the first record of a male dipteron possessing true mandibles. The hypothetical mandibles (Fig. 256h) of a dipteron are long, thin, sword-shaped structures fitted for piercing. They thus resemble the mandibles (md) of Tabanus (Fig. 255) and Culicoides (Fig. 253). They are situated between the clypeus, labrum-epipharynx, and max- illae, and are closely associated with the invaginations of the anterior 205] HEAD OF DIPTERA— PETERSON 35 arms of the tentorium. Structurally the hypothetical mandibles do not resemble the biting mandibles of the Orthoptera, but their situation and their association with the invaginations of the anterior arms of the tentorium are the same, which is far more important in determining their homology' than any particular form thej' may assume. The mandibles vary in their structure. In Psorophora (Pig. 251) they are long and needle-like, while in Tabanus, Culicoides, and the male of Simulium (Fig. 252) they are sword-shaped, and in Dixa (Fig. 254) spindle-like. The mandibles in the females of all species of Simu- lium (Fig. 250) examined are a trifle longer than those in the males (Fig. 252) and much broader at their distal ends. The greatest spe- cialization in structure and point of attachment with the head occurs in the long, thin, saw-like mandibles of Bibiocephala (Fig. 256) and Blepharocera. In these forms they are longer than the labium, blunt at the end, and toothed along the mesal margin, fitting against a similar edge on the lateral margin of the hypopharynx. All mandibles (md) of the Diptera are connected with the head- capsule cephalad of the maxillae (mx) and caudad of the labrum- epipharynx (1. ep) and the fronto-cljijeus (fr. c). In this respect they resemble the hypothetical type. In Psorophora, Dixa, Simulium, and Tabanus they are associated with the invaginations of the anterior arms of the tentorium. The proximal ends of the mandibles of Psorophora (Fig. 159) are bent cephalad, and articulate with the head-capsule at the distal ends of the crescent-shaped tentorial thickenings (t. th) which arise from the margins of the invaginations of the anterior arms of the tentorium. In Dixa (Fig. 254) the mandibles connect with the head- capsule at the ventro-caudal angles of the elypeus. An indistinct thick- ening extends dorsad from the point of articulation of each of the man- dibles toward the invaginations of the anterior arms of the tentorium. The mandibles of Simulium (Fig. 250 and 252) and Tabanus (Fig. 317) connect with the head-capsule directly ventrad of the invagina- tions of the anterior arms of the tentorium, but no direct connection occurs between them. In the female of Simulium the mandibles artic- ulate -n-ith a hook-shaped projection of the vertex. The mandibles of Tabanus (Fig. 255) are bifurcate at their proximal end and the lateral bifurcation articulates with the liead. The location of the mandibles of Bibiocephala (Fig. 155) and Blepharocera is generalized with respect to their position between the maxillae and the fronto-clj-peus, but their point of attachment with the head-capsule is decidedly specialized. They unite with chitinized pillars (Fig. 83) on the caudal aspect ventro-laterad of the invaginations of the posterior arms of the tento- rium. The proximal portion of eacb mandible is a long chitinized strip 36 ILLIXOIS BIOLOGICAL MONOGRAPHS [206- embedded in the membrane. These strips extend cephalad from their caudal connection to the cephalic margin of the membrane about the mouth-parts. At this point, M'here distinct tendons are attached, they turn abruptly ventrad and become free appendages. All connection between the mandibles and the invaginations of the anterior arms of the tentorium is lost. The relationship between the tentorium and the mandibles has not been observed in Culicoides for the lack of material. No other families of the Diptera outside of those to which the above- named genera belong, so far as observed, possess true mandibles or rudiments of the same. When mandibles are present, they are always of considerable size and probably functional. A number of investigators have described mandibles for many species not included in the above families. Langhoffer (1901) describes mandibles for the Dolichopodidae which are shown in this paper to be modifications of the epipharynx (Fig. 524 and .528). The apodemes of tlie muscids (Fig. 304, 308, and others) have been called mandibular tendons by MacCloskie and others. This is incorrect as shown by the figures and in the discussion of the maxillae. A number of workers (e.g., Wesche, 1909) believe that the mandibles have united with the labium and exist as chitinized strips on the cephalic aspect of the labium or as thickenings on the meson of the theca. Neither of these interpretations can be accepted when one takes into consideration the relative position of these so-called mandibles and the manner of devel- opment of the proboscis of the Calyptratae. The chitinized thickenings on the cephalic aspect of the labium are located caudad of the maxillae and the liypopharynx. This does not agree with the position of the mandibles of other insects. Furthermore, these thickenings are present in Tabanus where true mandibles occur. The chitinized thickenings on the meson of the theca in some of the Diptera can not be considered as rudiments of mandibles for many reasons. The most conclusive objection to tliis interpretation lies in the fact that these thickenings are very prominent in Simulium which has distinct mandibles in both sexes. When interpreting mouth-appendages, it is always necessary to take into consideration the generalized relationship between the mouth- parts and their association with the invaginations of the tentorium. It is also very desirable to observe a large series of forms before attempt- ing to homologize the parts. The above interpretations were apparently not made from either of tliese vantage-points. Maxillae. — All Diptera having functional mouth-parts have max- illae. They are, however, greatly reduced and modified in some genera,, and at first glance bear little or no relation to the structure or location 207] HEAD OF DIPTERA — PETERSOX 37 of the maxillae of generalized Diptera or other insects. Numerous intermediate stages of masillar}- development are present in the various species; consequently it is possible, and in fact comparativelj' easy, to trace thruout the order the main line of specialization and several side lines. The hypothetical maxillae of the Diptera (Fig. 257) resemble the maxillae of a generalized insect in their homologous sclerites, their posi- tion between the mandibles and the labium, and their close association with the invaginations of the posterior arms of the tentorium. Struc- turallj' they are composed of small triangular cardines (ca), long stipites (st), five-segmented palpi (mx.pl), needle-like galeae (g), and short laciniae (la). The cardines and stipites differ from those of gen- eralized insects in that they are embedded in the mesal membranous area ventrad of the occipital foramen. The palpi, galeae, and laciniae are free appendages. The proximal ends of the cai'dines are adjacent to the invaginations of the posterior arms of the tentorium. The struc- ture and position of the various parts of the hypothetical type have been traced thruout the order. The species in which the ptilinum is wanting are considered first. The cardines (ca) are small distinct triangular selerites in Trichoc- era (Fig. 260), Rhyphus (Fig. 261), Dixa (Fig. 262), and the female of Tabanus (Fig. 259). In these genera they are adjacent to the invagi- nations of the posterior arms of the tentorium. The cardines of Simu- lium (Fig. 258), in both males and females, differ from those of the above genera in that thej' are large and occupy nearly all of the mem- branous area between the postgenae dorsad of the stipites. Their margins are also somewhat indistinct. No other forms figured have distinct sclerites that are homologous with the cardines of the hypo- thetical type. The maxillae of Rhabdophaga (Fig. 268), Bibiocephala (Fig. 269), and Chironomus (Fig. 270) connect with the invaginations of the posterior arms by means of narrow chitinized processes which arise from the stipites proper. Undoubtedly these pieces are reduced cardines which have lost the suture that separates them from the stipites. The presence of this suture is suggested by the suture-like depression in the male of Bibioeepluila (Fig. 76). Excepting Promachus (Fig. 276) and the above forms, the cardo is wanting in all the maxillae figured. The maxillae of Psychoda (Fig. 263) and Sciara (Fig. 267) closely resemble some of the above maxillae, but the cardines as chi- tinized pieces are apparently wanting. There is a distinct membranous area between the proximal ends of the stipites and the invaginations of the posterior arms of the tentorium. From forms such as these it is concluded that tlie cardines have been lost as chitiidzed areas. No other interpretation seems possible with the evidence at hand. 38 ILLINOIS BIOLOGICAL MONOGRAPHS [208 The stipites (st) are of various shapes and sizes as can be seen in the figures. In Rhabdophaga (Pig. 268), Bibiocephala (Fig. 269), Chironomus (Fig. 270), and possibly Mycetobia (Fig. 90), they have united to form a chitinized strip or plate in the membranous area dorsad of the labium. This piece should not be confused with the submeutum of the labium. In all species in which the postgenae have not united ventrad of the occipital foramen, the proximal ends of the stipites are near the invaginations of the posterior arms of the tentorium. In all species where the postgenae form a continuous plate, the stipites are reduced in size and situated at or beyond the ventral margin of the head, as in Mydas (Fig. 319) and Eristalis (Fig. 328). In other words, the usual association between the maxiUae and the invaginations of the posterior arms has been lost. Psilocephala (Fig. 281) and Psorophora (Fig. 96) are exceptions to the last statement. In Psilocephala chi- tinized thickenings (ch. th) are present on the ental surface of the postgenae ventrad of the occipital foramen, and these are undoubtedly rudiments of the stipites. The stipites of Psorophora (Fig. 266 and 96) are long, free rod-like structures located entad of the postgenae. They extend between the occipital foramen and the ventral margin of the head. The stipites of Geranomyia (Pig. 382) and Limnobia (Fig. 386) are also entad of the postgenae. In these genera their proximal ends are united and they have no connection with the head-capsule. The stipites of Tipula (Fig. 277) resemble those of Geranomyia and Limnobia, but there is greater reduction in size, and they are completely united along their mesal margins, thus forming a single median piece. The maxillae of Promachus (Fig. 84) differ from those of all other genera in that the stipites and the cardines are united on the meson and continuous with the postgenae near the occipital foramen. Narrow membranous areas separate tlie maxillae from the postgenae near the ventral margin of the head. This unique modification of the maxillae agrees with the striking modifications in the other mouth-parts. The figures show the variations in other genera belonging to this group. In general it can be said that the stipites have been modified by reduction and by removal to the ventral margin of the head and in some cases are even located on the basiproboscis. The maxillary palpi (mx. pi) of the Nematocera figured have from two segments — Geranomyia (Fig. 382) and the female of Psorophora (Pig. 266) — to five segments. The usual number is four or five. In the Brachyeera only one articulating segment is present. This segment in Tabanus (Fig. 259) connects with an elongated portion of the stipes which is called the palpifer by some. In this study the palpifer is considered as wanting, since no palpus of the Diptera possesses over 209] HEAD OF DIPTERA — PETERSON 39 five segments and furthermore no piece is present at the base of any generalized palpus which can be homologized with the palpifer of gen- eralized iusects. The greatest reduction in the palpus of the Nematocera occurs in Geranomyia (Fig. 382), while in the Bi'achj-cera the palpus of Mydas (Fig. 271) is a mere lobe. A small finger-like structure arises from the ventro-mesal margin of each stipes and pi-ojects mesad to the caudal aspect of the hypo- pharynx in Tabanus (Fig. 259) and Simulium (Fig. 258). These pieces are apparently homologous with the laciniae (la) of generalized insects. The distal ends of these projections articulate against the caudal aspect of the hypopharjTix (Fig. 496 and 497), and in this respect they differ from the laciniae of generalized insects. These pieces in Tabanus have been described as laciniae by Patton and Cragg (1913). A distinct lobe is present mesad of the palpus in the majority of the Diptera that do not have a ptilinum. This structure is unquestion- ably the galea (g), for in specialized insects which possess a distinct galea the lacinia is generally reduced in size and in some cases wanting. This tendency of development prevails in the Diptera. If the above pieces in Tabanus and Simulium which are described as laciniae are truly such, there can be no question regarding this interpretation of the lobe adjacent to the palpus. The galeae vary considerably in size and shape. They are long and needle-like in Tabanus (Fig. 259), in the female of Psorophora (Fig. 266), and in Empis (Fig. 274), Exo- prosopa (Fig. 285), and Eulonchus (Fig. 284a); while in Trichocera (Fig. 260), Disa (Fig. 262), Sciara (Fig. 267), Bittacomorpha, Chi- ronomus (Fig. 270), Lonchoptera (Fig. 280), Sceuopinus (Fig. 282), and the male of Psorophora (Fig. 266) they are greatly reduced. In Bibio (Fig. 264) and Geranomyia (Fig. 382) they are mere rudiments. They are wanting in Ehabdophaga (Fig. 268), Tipula (Fig. 277), Helobia (Fig. 385), Aphioehaeta (Fig. 278), Pipunculus (Fig. 279), Platypeza (Fig. 272), and Dolichopus (Fig. 284). The development of the maxillae of the genera possessing a ptilinum will now be considered. No cardines or laciniae are present in this group. The maxillary palpi are one-segmented and are present in aU forms except Conops (Fig. 305). The palpi interpreted here as maxil- lary palpi have been called labial palpi by some (e.g., "VVesche, 1909). The stipites and galeae are present in aU the species studied, and they undergo decided morphological changes. All connection or association between the maxillae and the invaginations of the posterior arms of the tentorium has been lost. This loss is even more pronounced than in the Brachycera, since in all but a few species figured the maxillae are far removed from the head and situated near the distal end of the 40 ILLIXOIS BIOLOGICAL MOXOGRAPHS [210 well-developed basiproboscis. This migration of the maxillae in the Cyelorrhapha has not altered their generalized position between the labrum-epipharyux and the labium. The stipites of genera having a ptilinum show all stages of in- growth from a turned-in fi-ee edge or end (st-e), to forms in which it is entirely entad of the membrane of the basiproboscis, as in Musca. Eristalis (Fig. 286), Eulonclms (Fig. 284a), and Exoprosopa (Fig. 285) are the only forms without a ptilinum wliich show an ental growth of the stipites. These genera make a good starting point for explaining the characteristic development found in the Acalyptratae and the Calyp- tratae. The following scheme of lines and dots has been adopted on the drawings in order to show the degree of ingrowth of the stipes. A continuous solid line on the stipes indicates a definite ectal boiindary which connects with the membrane of the basiproboscis. A broken line indicates an ental edge or end which is free of the membrane between it and the observer. The membrane is represented by stippling. For convenience of description and homology the following division of the stipes has been made : st represents the ectal portion of the stipes and st-e the ental portion ; and st is further divided into st-1 and st-2 as seen in Coelopa (Fig. 288). In Exoprosopa (Fig. 285) and Bulonchus (Fig. 284a) the proximal end of the stipes is free and entad of the membrane, while the cephalic edge and the dorsal end are entad in Eristalis (Fig. 286). From a form similar to Eristalis it is possible to develop a stipes which would resem- ble that of Sepsis (Fig. 287), Coelopa (Fig. 288), and Calobata (Fig. 296). In Sepsis the palpus is greatly reduced, but it connects with an ectal portion of the stipes (st) which in turn gives rise to the free ental portion (st-e). The free ental part extends ventrad and is continuous with the galea, which emerges from the membrane near the base of the labrum as a free appendage. The stipes of Coelopa (Fig. 288), Sapro- myza (Fig. 289), and Sphyracephala (Fig. 293) is similar to that of Sepsis, but in these forms the palpus arises from the cephalic margin of the basiproboscis. The palpus is connected with the stipes proper by means of a long chitinized strip (st-1) which is usually covered with setae. This ectal poi'tion of the stipes (st-1) is present in all but a few genera, such as Chloropisca (Fig. 306), Heteroneura (Fig. 298), Chyro- mya (Fig. 299), Loxocera (Pig. 300), and Euaresta (Fig. 292). In a number of forms, particularly in the Calypti-atae, a small chitinized area is present ventrad of the palpus. This is regarded as a secondary chitinization. The ectal portion of the stipes (st-2) is present in a majority of the Acalyptratae and in one or two of the Calyptratae. The ental portion of tlie stipes (st-e) is always present in the members of this groiip. In Desmometopa (Fig. 303), Chloropisca (Fig.. 211] HEAD OF DIPTERA—PETERSOX 41 306), Conops (Fig. 305), and the Calyptratae it has no connection with the ectal portion of the stipes (st-2) or the membrane, and by many writers is commonly called the apodeme. The free so-called apodeme is unquestionably derived from the ental ingrowth of the stipes, as illustrated by the modifications found in the following genera : Coelopa (Fig. 2S8),'Sapromyza (Fig. 289), Tetanocera (Fig. 297), Archytas (Fig. 309), Musea (Fig. 304), and others. The development of the galea (g) maj' be traced thruout the Cy- clorrhapha in a manner similar to that of the stipes. In Eristalis (Fig. 286) the galea is a long free appendage arising from the ventral end of the stipes near the proximal end of the labrum-epipharynx. Its length and size are greatl.v reduced in Sepsis (Fig. 287), but its position is identical with that of Eristalis. Thruout the majorit.v of the Acalyp- tratae the position of the galea resembles that of Sepsis. Its size and form undergo some change, as can be seen in the figures. In the Calyp- tratae and some of the Acalyptratae the galea articulates with the proximal end of the labrum and is more or less firmly connected with the same. The ectal exposure of the galea is very small in these forms. The large galea of the Acalyptratae has been considered as the maxillary palpus by Wesche (1902). This interpretation is highly improbable. Lahiiim. — The labium is the most specialized and characteristic appendage of the mouth of Diptera. Its structural modifications are very striking among the specialized genera, such as the Cyelorrhapha. These modifications are largely due to the reduction of the parts and the excessive development of membranous areas, and they agree with similar types of modification in other head- and mouth-parts. To explain the imique development of the labium of Diptera, it has been necessary to make a critical study of the generalized condition of this appendage as it occurs in the Nematoeera and to compare it carefully with the labia of more generalized insects. As is well known, the labium of a generalized insect is the posterior, independent, flap- like mouth-part, made iip of a submeutum, meutum, and ligula. The ligula is further divided into palpigers, palpi, paraglossae, and glossae. The labium of a generalized dipteron resembles that of a generalized insect in its caudal position and in its independent condition, but it is very different in structure. It is more or less enlarged and not flat and flap-like, and the palpi and palpigers are always wanting, so far as observed. Since the position of the palpi and the palpigers is very useful in orienting the sclerites of the labium of generalized insects, their absence in Diptera makes it exceedingly difficult to homologize cor- rectly and locate the submentum, mentum, and the parts of the ligula. The membranous condition of the labium also adds to this difficulty. 42 ILLINOIS BIOLOGICAL MONOGRAPHS [212 In order to get some light on this problem, a study was made of the labium, particularly the submeutum and mentum, of a number of generalized insects of the more common orders. The literature of this subject was examined, but no satisfactory results were obtained from this source. After a careful study of a number of labia, the following general characteristics which bear upon the labium of Diptera, were noted. The submentum is the large proximal segment, while the mentum is usually small and in some cases very thinly chitinized and almost obsolete. The sutures separating the mentum from the submentum and the ligula are only represented by small remnants in Melanoplus. The ligula, so far as observed, comprises the movable parts of the labium, while the mentum and submentum are more or less firmly united with the head-capsule. The proximal part of the ligula is usually well de- veloped and gives rise to the palpigers, palpi, paraglossae, and glossae. The glossae are located between the paraglossae, and in a number of forms a distinct depression or thickening extends proximad between the glossae and the proximal margin of the ligula. With these observations as a basis for comparison, the labium of such generalized Diptera as Chironomus (Fig. 371), Simulium (Fig. 366), Trichocera (Fig. 365), Dixa (Fig. 375), and others may be inter- preted as follows. The mesal membranous area of the caudal aspect of the head, which is bounded by the postgenae (po), the occipital foramen (o. f), and the proximal chitinized piece of the labium (the), is made up of the submentum, mentum (su. me), and the cardines (ca) and stipites (st) of the maxillae (mx). Since this area is largely mem- branous, it is impossible to determine the boundaries of these sclerites. The areas laterad of the cardines and the stipites apparently belong to the maxillae, while the area mesad of these parts is made up of the submentum and mentum (su. me). The important feature concerning this mesal membranous area is the fact that the maxillae and the labium both pla.y a part in its formation. This undoubtedly indicates that the submeutum and mentum, of a more or less fixed nature in generalized insects, have been more extensively fixed in the Diptera, and that the submentum and mentum are included in the membrane developed from the stipites and cardines. Such an interpretation is altogether possible, since the proximal portions of the maxillae are adja- cent to the submentum and mentum in generalized insects. The ligula (Ig) of the generalized Diptera agrees with the ligula of generalized insects in that it is the movable part of the labium. Structurally it is composed of a well-developed proximal area which gives rise to two large bulb-like paraglossae (pgl) and to two small 213] HEAD OF DIPTERA— PETERSON 43 membranous glossae (gl) wliicli are located between the paraglossae. The palpigers and labial palpi are wanting, but if in the future some form is discovered wliieh shows these structures, thej' will undoubtedly be found on the area liere described as the ligula. The proximal portion of the ligula has a decided furrow or thickening on its caudal aspect along the meson. This thickening is characteristic of a number of Diptera and resembles the proximal portion of the ligula of a number of generalized insects. This mesal thickening marks the line of fusion of the two parts of the labium during embryonic development. The above interpretation of the labium is on the whole very satis- factory for the numerous modified types found in the various families of the Diptera, and with this interpretation it is possible to formulate a hj'pothetical labium. This has been done in this study; but there have been added to this labium the early stages of development of the more important secondary structures Mhich are characteristic of the labia of Diptera. It will therefore be advisable to call such a hypothetical labium a tj-pical labium in order to distinguish it from the true hypothetical type of other parts of this study. A typical labium of the Diptera (Fig. 1, 73, 140, 362, and 363) is made up of a submentum, mentum, and ligula. The submentum and mentum (su. me) are firmly united with the head and constitute the greater portion of the mesal membranous area of the caudal aspect of the head. The ligula (Ig) is the large swollen and movable portion of the labium and consists of the mediproboscis (mpr) and the disti- proboscis (dpr). The mediproboscis has a chitinized area on its caudal aspect which is commonly called the theca (the). The distiproboscis is composed of two large membranous bulb-like paraglossae (pgl) and two small membranous glossae (gl) which are located between the proximal parts of the paraglossae. The important and characteristic features of a typical labium are the chitinized pieces on the caudal and lateral as- pects of the paraglossae and the trachea-like structures on the mesal aspects. The details of the various parts will be more fully discussed as each part is considered and its modification traced thruout the order. The submentum and mentum (su. me) are present as a membranous area in a majority of the Nematocera and in the females of Tabanus (Fig. 74). This area undergoes considerable modification, as was seen iu the discussion of the maxillae and postgenae, and is illustrated by the figures. Rhyphus (Fig. 80 and 374) is apparently the only genus which has within this area a chitinization which can not be considered as a modification of the maxillae or of the postgenae. This piece is a more or less distinctly chitinized, inverted-flask-shaped area between the maxillae. If this is a primary chitinization, it is probably a rem- 44 ILLINOIS BIOLOGICAL MONOGRAPHS [214 nant of the submentum. A similarly situated area found in Mycetobia has been homologized by some writers with that of Rhyphus. This interpretation is undoubtedly incorrect, since this area in Mycetobia (Fig. 90) gives rise to chitinized projections at its ventro-lateral angles and these in turn connect with the maxillary palpi and the galeae. Furthermore, the relationship which this piece bears to the proximal end of the theca (the) would tend to disprove such an interpretation. This piece in Mycetobia is undoubtedly a specialization of the maxillae similar to the modifications found in Bibiocephala (Fig. 83) and Rhab- dophaga (Fig. 86). In all genera where the postgenae have grown together on the meson the submentum and mentum have been elimi- nated, unless one regards the area between the ventral margin of the head and the theca as derived from these areas. This area, as already described for the Cyclorrhapha, is very extensive and forms the caudal portion of the basiproboscis (bpr). The proximal portion of the ligula or mediproboscis (mpr) of the tj'pieal labium is largely membranous, but it has ou its caudal aspect a distinctly chitinized area, the theca (the), which has a distinct furrow on its meson. The shape, size, and degree of chitinization of the theca vary greatly, as can be seen in Bibio (Fig. 364), Triehocera (Fig. 365), RhjT)iius (Fig. 374), Promachus (Fig. 376), Tabanus (Fig. 391), Chyromya (Fig. 411), Conops (Fig. 420), Rhamphomyia (Fig. 424), and Musca (Fig. 466). There is a distinct furrow or thickening on the meson of the majority of the Nematocera and the Brachycera, and rem- nants of these thickenings occur also among the Cyclorrhapha. In some of the Diptera the structural condition of the meson has a marked influ- ence on the shape of the dorsal and ventral margins of the theca. The cephalic aspect of the proximal portion of the ligula of a typical labium is concave and membranous and connects with the proximal part of the lance-like portion of the hypopharynx. In the Nematocera the cephalic aspect resembles the typical labium, and in the Brachycera and in a majority of the Cyclorrhapha it has a distinctly chitinized groove. This is well "illustrated by Tabanus (Fig. 392), Eristalis (Fig. 441), and a majority of the Calyptratae. The degree of chitinization varies con- siderably, and in some forms heavy, chitinized, cord-like pieces extend along the sides of the groove from the glossae to the proximal end of the labium. The distiproboscis of the typical labium is composed of two large independent, highly membranous, bulb-like paraglossae (pgl), usually called oral lobes or labellae, and two small membranous glossae (gl). Each paraglossa has on its lateral and caudal aspects a Y-shaped chi- tinized support which has been commonly called the furca. For con- 215] HEAD OF DIPTERA — PETERSON 45 venienee in description and as an aid in tracing the development of the parts of the furca thruout the order, it has been divided into furca-1, which is the stem of the Y, furca-2, which is the dorsal arm of the Y, and furea-3, which is the ventral arm. The furca articulates with a small sclerite which is located between the proximal end of furca-1 and the distal end of the chitinized furrow on the meson of the theca. This piece has been called the sigma (si). Another small, independent sclerite is located in the membrane just laterad of the sigma and this may be kno'mi as kappa (k). Eacli paraglossa has on its mesal aspect two trachea-like structures which arise from the proximal portion of the glossa. These structures are commonly called pseud otracheae (ps). A general survey of the characteristics of the paraglossae of the various labia shows that they are usually bulb-like, membranous, and somewhat flexible. In these respects they diifer decidedly from the firmly chitinized, flap-like labia of many generalized insects. Their size and shape vary greatly, as can be seen in Bibio (Fig. 364), Leia (Fig. 368), Promachus (Fig". 376), Geranomyia (Fig. 382), Tipula (Fig. 384), Tabanus (Fig. 390), Conops (Fig. 417), Empis (Fig. 421), Siphona (Fig. 458), Musca (Fig. 467), Stomoxys (Fig. 479), and Olfersia (Fig. 488). The use to which the labia are put seems to have some influence on their form. The main line of development thruout the genera figured is toward the type found among the Calyptratae, in which the labia are usually large, decidedly membranous, and joined together on the dorso-caudal areas, as in Hydrotaoa (Fig. 475), Sarcophaga (Fig. 477), Sepsis (Fig. 439), Loxocera (Fig. 461), Tetanoeera (Fig. 463), and many other genera. The membranous development of the paraglossae is not always a good indication of the main line of specialization. In a number of scattered genera, Chironomus, Rhjphus, Aphiochaeta, Chloropisca, Platypeza, Leptis, Psilocephala, and Lonchoptera, it is next to impossi- ble to make out the chitinized pieces, such as kappa, sigma, and furca, because of the membranous condition of the entire labium. Outside of the above-named forms, the chitinized pieces of the paraglossae are usually distinct when present. These supports may be secondary iu origin or they may be remnants of former chitinized parts of the para- glossae. It is possible to show how the various cliitinized pieces of the majority of the labia may have been developed from the typical form. The sclerite designated as kappa (k) on the typical labium is onlj' present in Tabanus (Fig. 390 and 391), Tipula (Fig. 388), and Bitta- comorpha (Fig. 85). No otlier dipteron gives any evidence whatever of such a sclerite. In the above-mentioned genera the pieces are em- bedded in the membrane laterad of the ventral ends of the theca. Some 46 ILLINOIS BIOLOGICAL MONOGRAPHS [216 one has interpreted these pieces as rudimentary palpigers or palpi. This may or may not be correct. It is possible for palpi to be in such a position ; but since no other genera have similar pieces, and since they are so decidedly dissimilar to the labial palpi and palpigers of general- ized insects, they are here regarded as secondary sclerites. The sclerite designated as sigma (si) is present as a chitinized thick- ening at the ventral end of the theca, as in Eristalis (Fig. 443), or as a distinct piece, as in a majority of the Brachycera and the Cyclorrha- plia. In aU genera it is situated between the ventral margin of the theca and the furca. Only a few genera of the Nematocera, such as Tipula (Fig. 388) and Psorophora (Fig. 380), have these sclerites. They undergo some modification in size and structure as can be seen in the following genera: Tabanus (Fig. 391), Mydas (Fig. 397), Conops (Fig. 418), Borborus (Pig. 437), Eristalis (Fig. 443), Coelopa (Fig. 448), and Scatophaga (Fig. 470). The furca of Bibio (Fig. 315) and that of Tabanus (Fig. 317) closely resemble the typical form. In Bibio, furca-1 (f-1) and furca-2 (f-2) are one continuous piece, while furca-3 (f-3) is a distinct arm. In Tabanus, furca-2 and furca-3 are distinctly chitinized areas arising from the distal end of furca-1. Only one chitinized support is present in Sciara (Fig. 314), Rhabdophaga (Pig. 313), Psychoda (Fig. 318), Stratiomyia (Fig. 331), and Trichocera (Fig. 311). In Trichocera this support has a decided dorsal bend near the constriction of the para- glossae. This bend is also present in Psychoda and Stratiom^'ia, but the constriction is wanting. The distal portion of the furca beyond the bend is homologous with furca-2, and furca-3 is wanting in these forms. Furca-2 is present and furca-3 is wanting in Scenopinus (Pig. 325) ; furca-3, however, is present in more species than furca-2. Such is the case with Borborus (Fig. 342), Chrysomyza (Pig. 341), Coelopa (Pig. 337), Tetanocera (Fig. 344), Scatophaga (Pig. 357), Musca (Pig. 351), and Thelaira (Pig. 346). Furca-1 (f-1) varies considerably thruout the order. In general- ized forms where the dorso-caudal portions of the paraglossae are not joined together the fureae are always well separated. They are also separated in some forms where the paraglossae are joined, as in Mydas (Fig. 397) and Eristalis (Pig. 443). In Chyromya (Fig. 411), Dro- sophila (Pig. 454), Tetanocera (Fig. 463), and Sepsis (Fig. 439), an intermediate piece joins the mesal ends of furcae-1 while in Sarcophaga (Pig. 477), Musca (Pig. 466), Coelopa (Pig. 448), Sapromyza (Pig. 409), Chrysomyza (Pig. 457), Heteroneura (Pig. 459), and Oecothea (Pig. 452) furcae-1 are united and form one continuous U-shaped piece. This type of fureae is present among the Calyptratae. The fureae of 217] HEAD OF DIPTERA — PETERSON 47 specialized forms, such as Olfersia (Fig. 488), Couops (Pig. 418), Siphona (Fig. 355), Empis (Fig. 421), and others, are not differen- tiated, since the greater part of the lateral aspects of the paraglossae is chitinized. In the typical labium two simple tracliea-like structures, commonly known as pseudotracheae (ps), arise from the proximal part of each glossa and extend onto the mesal membranous aspect of each paraglossa. These trachea-like structures are in reality small chitinized troughs which serve as conduits for the liquid food. Pseudotracheae are unique structures and peculiar to Diptera, so far as known. They are present in only a few generalized forms, but from these genera it is possible to develop the pseudotracheal arrangement and structure of the more specialized Diptera. It is consequently as.sumed that the pseudotracheae have probably arisen only once within the order, and that this happened some time after the group as a whole -was set off as a distinct order. The psedotracheae (ps) of Tipula (Fig. 383) resemble those of the typical labium in that the two main pseudotracheae arise from each glossa and extend over the mesal membranous area of the paraglossa, one of the pseudotracheae extending caudad and the other cephalad. These ducts are secondarily branched and resemble a fern. The pseudo- tracheae of Mycetophila (Fig. 11) and Leia (Fig. 368) are reduced and only the caudal pseudotracheae extend over the paraglossae. The para- glossae in these genera are united along the meson and form a single large lobe. The cephalic pseudotracheae are indicated by small rudiments in Mycetophila (Fig. 11). The pseudotracheae in these forms resemble the typical labium in that they are simple, unbranched, chitinized troughs. From the typical labium, or from the p.seudotracheae as they occur in Tipula, it is possible to derive the arrangement and structure of the pseudotracheae as they are found in Tabauus (Fig. 390) and similar forms, where two long pseudotraclieal trunks (m. ps) extend cephalad and caudad from the .glossae (gl) and give rise to many branches on their ventral side. These branches extend ventrad over the entire mesal area of the paraglossa (pgl). The arrangement of the pseudotracheae of most Diptera is readily derived from a form similar to Tabanus. The arrangement in Scenopinus (Pig. 400), Psilocephala (Fig. 403), and many of the Calyptratae resembles that in Tabanus. In such genera as Stratiomyia (Fig. 396), Oecothea (Fig. 453), Coelopa (Fig. 449), and Heteroneura (Pig. 460) no main collecting duets (m. ps) extend beyond the glossae. In many genera, such as Chloro- pisca (Fig. 431) and Chyromya (Fig. 412), no line of demarkation can be drawn between the proximal ends of the pseudotracheae and the glossae. 48 ILLINOIS BIOLOGICAL MONOGRAPHS [218 U-shaped or open ring-like thickenings are present in the pseudo- tracheae of the more specialized Diptera. They do not occur in the simple pseudotracheae of Mycetophila or in some of the highly special- ized forms. The histological structure of a pseudotrachea has been clearly demonstrated by several workers. According to Dimmock, "The pseudotracheae on the inner surfaces of the labellae of Musca are cylin- drical channels, sunk more or less deeply iuto the surfaces of the labeUae according to the amount that that surface is inflated, and they open on the surface in zig-zag slits. These channels are held open by partial rings, more strongly chitinized than the rest of the membrane of the cylinder. As seen from above in Musca, [Fig. 485], the pseudotracheae appear to be supported by partial rings, one end of each of which is forked The pseudotracheae of Eristalis are so nearly like those of Musca [Calliphora] vomitoria that I have not figured those of the former." All my observations of the histological structure of pseudotracheae agree with those made by Dimmock. Tho no attempt was made to work out the detail of the liistological structure in the various genera studied, a number of interesting facts were observed. The chitinized, taenidia-like thickenings (ps. th) in Ochthera (Fig. 445 and 483) are large U-shaped structures which are partially embedded in the membrane. The ends of these thickenings project considerably beyond the surface of the membrane and resemble these structures in Bombylius major (Fig. 482), as figured by Dimmock. The pseudo- Ira cheae of Calobata (Fig. 446) have developed into rows of small chi- tinized teeth (tee). The pseudotracheal area of the paraglossae undergoes its greatest specialization in forms in which the paraglossae assume a biting func- tion. This biting type is brought about by the development of distinct chitinized teeth arising between the proximal ends of the pseudotracheae. Rudimentary or well-developed teeth occur in Musca (Fig. 467), Sar- cophaga (Pig. 478), Seatophaga (Fig. 472), Lispa (Fig. 481), and Stomoxys (Fig. 480). In Musca the small, chitinized, so-called pre- stomal teeth (tee) are present between the proximal ends of the pseudo- tracheae. In Seatophaga and Lispa these teeth are large and distinct. Their greatest development occurs in Stomoxys, and so far as observed pseudotracheae are wanting in this form. An extensive discussion of the development and the structure of the chitinized teeth of the para- glossae has been given by Patton and Cragg (1913). The glossae (gl) of a typical labium (Fig. 1 and 73) are two small lobes located between the proximal portions of the paraglossae distad of the furrow on the theca and at the distal end of the cephalic groove. Thruout the order the glossae are between the paraglossae and at the 219] HEAD OF DIPTERA— PETERSON 49 distal end of the cephalic groove. They are not well-defined structures in all labia. In ChLfonomus (Fig. 371), they are two small membranous lobes, while in Simulium (Fig. 366), Rhabdophaga (Fig. 367), Bibio (Fig. 364), and Rhyphus (Fig. 374) they have the form of a single median membranous lobe. The glossae of Simulium are of particular interest since they have a great number of minute chitinized thickenings which radiate from the proximal end. So far as known these thicken- ings bear no relation to the psedotracheae of the paraglossae. The glossae of Tabanus (Fig. 391) are united and form a chitinized triden- tate piece with the median tooth the longest. The glossae of Lonchop- tera (Fig. 407) illustrate a form intermediate between a median spine, such as occurs in Psorophora (Fig. 381), Aphiochaeta (Fig. 393), Empis (Fig. 422), and Exoprosopa (Fig. 426), and the U-shaped structure characteristic of the Cyclorrhapha. The glossae of the Calyptratae re- semble in general the glossae of Musea (Fig. 465). In the genera of this group the cephalic ends of the U-shaped piece are free and project cephalad from the point of attachment of the pseudotracheae. The glossae are not well defined in a few genera, Sapromyza (Fig. 410), Chyromya (Fig. 412), and Chloropisca (Fig. 431), for example, and it is impossible to differentiate the glossae from the chitinized groove of the mediproboscis and the proximal ends of the pseudotracheae. The glossae of Promachus (Fig. 379) are specialized in that they give rise to two thickenings which extend dorsad in the groove of the labium and serve as guides for the hypopharynx and galeae. EPIPHARYNX AND HYPOPHARYNX The anterior end of the alimentary canal of the Orthoptera and of insects in general is divided transversely into two parts, one forming the cuticular lining of the clypeus and labrum and the other the lining of the opposite side of the mouth ca^^ty. The portion lining the clypeus and labrum is known as the epipharynx (ep), and that of the opposite side as the hypopharynx (hp). Each lining may be subdivided into several parts. These are of particular significance in a study of the epipharj'nx, which has a distinct chitinized mesal piece, and two lateral chitinized pieces which are situated near the clypeo-labral suture. These lateral pieces, which have been designated as tormae (to), and, so far as I know, are described here for the first time, project cephalad toward the clypeo-labral suture in Melanoplus (Fig. 515) and Gryllus (Fig. 516) and connect with both the labrum and clypeus. In Gryllus they are interpolated between the clypeus and the labrum and appear as small triangular sclerites on the cephalic aspect. The tormae of Peri- planeta (Fig. 514) are not as well developed as in the above-named 50 ILLIXOIS BIOLOGICAL MONOGRAPHS [220 genera, but they are present and project toward the cephalo-lateral corners of the labruni. The caudal end of the epipharynx in many in- sects gives rise to long chitinized arms which have been called cornua (eu). The hypopharynx may be subdivided into a distal, unpaired, me- dian piece, which is usually called the hypopharynx, and a proximal paired area. The chitinized portion of the anterior end of the alimentary canal of Diptera can be homologized with the epipharynx and the hypophar- ynx of generalized insects. The following hypothetical epipharynx and hypopharynx (Fig. 493) and their closely associated parts have been constructed for Diptera. In the figures of tlie lateral views of the hypothetical type an enlarged, three-sided, chitinized tube extends cau- dad from the dorsal end of the hypopharynx and epipharynx. It has been called the oesophageal pump (oe. p). This is not a part of the epipharynx or of the hypopharynx, but is a modification of the pharynx, a portion of the alimentary canal. All of the chitinized parts ventrad of the membranous area at the cephalic end of the oesophageal pump belong to the epipharynx and the hypopharynx. The dorsal ends of the epipharynx and the hypopharynx are united and form a single chi- tinized tube, and this has been called the basipharynx (bph). Except for this union, the epipharynx and the hypopharynx are continuous chitinized pieces with lance-like distal ends. The distal portion of the epipharynx is joined to the labrum by a membrane along its lateral margin. The tormae in the hypothetical type project from the lateral margins of the epipharynx and unite with the latero-ventral portions of the fronto-clypeus (fr. c). Two projections occur at the dorsal end of the basipharynx, and these are considered homologous with the cor- nua (cu) of the epipharynx of generalized insects. The distal end of the hypopharynx is a free lance-like organ, and a salivary duct (s. d) enters its proximal end just dorsad of the place where it joins the labium (li). The salivary duct extends thru the hypopharynx to its distal end. The oesophageal pump of the alimentary canal is closely associated with the epipharynx and hypopharynx in all the Nematocera and in Promachus (Fig." 517), Tabanus (Fig. 494), Leptis (Fig. 520), and Psiloeephala (Fig. 533) of the Brachyeera. In a majority of the above forms, the oesophageal piimp is an elastic, semi-chitinized, three-sided tube with muscles connecting with each of its surfaces. A contraction of these muscles expands the tube, which \ipon their relaxation assumes its normal shape. In some forms, as Tabanus and Promachus, there is only one chitinized elastic surface. In a number of genera, as Chi- ronomus (Fig. 531), Psychoda (Fig. 529), and Leptis (Fig. 520), the 221] HEAD OF DIPTERA— PETERSON 51 tube is more or less membranous and not distinctly three-sided. Tlie oesophageal pump is wanting in all the Diptera except those named, and the membranous oesophagus connects directly with the basipharynx. The oesophageal pump shows considerable variation in its shape, posi- tion, and size, as can be seen in the figures of Bibio (Fig. 523), Rhyphus (Pig. 508) and others. The basipharynx (bph) is interpreted as including all of the united portions of the epipharynx and the hypopharynx, but the extent of this union varies somewhat in the different genera. In a majority of the Nematocera no sutures or constrictions occur between the basiphar- ynx and the lance-like portions of the epipharynx and the hypophar- ynx. Such constrictions and secondary sutures do occur in a majority of the Brachycera, as in Leptis (Fig. 520) and Promaehus (Fig. 517), and in all of the Cyclorrhapha. The basipharynx (bph) varies in size and shape, as can be seen in the figures. Muscles connect with the cephalic and caudal aspects of the basipharynx, those on the cephalic aspect expanding the basipharynx and thus producing suction. This sucking apparatus is well developed in all forms which have no oesophag- eal pump. The chitinized projections at the dorsal end of the basiphar- ynx, called the cornua (cu), vary in shape and size. Some are blunt, others long and narrow, as in Leptis and the Calj^ptratae, and still others are disk-shaped, as in Promaehus (Fig. 517). Distinct tormae (to) are present in Diptera except in a few species of the Nematocera. In all the Nematocera and in Leptis (Fig. 520), Psilocephala (Pig. 533), Platypeza (Pig. 543), Aphiochaeta (Fig. 544), Lonchoptera (Fig. 539), and Scenopinus (Fig. 538), they resemble the hypothetical type in that they join with the fronto-clypeus. In other genera the tormae have an exposed portion located ventrad of the fronto-clypeus and all connection between the fronto-clypeus and the tormae is lost, except in Simulium (Fig. 497) and Tabanus. The variations in the shape and the extent of the tormae is well illustrated by the numerous figures. The so-called fulcrum described by numerous morphologists for the Calyptratae is composed of the tormae and the basipharynx. A more or h^ss distinct secondary sutiire (s. s) is shown in tlie drawings as separating the tormae from the basipharynx, and the broken line on the tormae indicates the place of connection of the membrane of the basiproboscis with the tormae. In figures of the Nematocei'a and of forms in which the tormae connect with the fronto- clypeus the broken line indicates the place of union between these parts. The epipharynx (ep) is present and closely associated with the labrum in all Diptera having functional mouth-parts. The interrela- tionship between the epipharj'nx and the labrum has been discussed 52 ILLINOIS BIOLOGICAL MONOGRAPHS [222 under the heading labrum. The epipharynx in a number of generalized Diptera, such as Tabanus (Pig. 494), Simulium (Fig. 497), Dixa (Fig. 501), Limnobia (Fig. 507), and Sciara (Fig. 513), resembles the hypo- thetical type. In the majority of the Diptera it differs from the hypo- thetical type in that it is completely separated from the basipharynx by a constriction or a secondary suture. This hinge in the epij^harynx permits the proboscis to bend at this point when it is withdrawn into the oral cavitj\ The lance-like portion of the epipharynx in the Calyp- tratae and some other forms is completely separated from the basiphar- ynx by the development of a special piece which is commonly called the hyoid (hy). The lance-like portion of the hypopharynx also articu- lates against the hyoid. The hyoid is a secondary sclerite which origi- nated from the epipharynx or the hypopharynx and serves the purpose of keeping open the alimentary canal, which passes thru it. A structure similar to the hyoid of Musca (Fig. 600) is found in Stomoxys (Fig. 599), where a large and strong trachea-like tube extends between the dorsal ends of the lance-like portions of the epipharynx, the hypophar- ynx, and the basipharynx. In size and shape the epipharynx agrees more or less closely with the labrum. The epipharynx in sucking Diptera is, as a rule, long and needle-like, while in other forms it is usually short and blunt. In many genera of the Acalyptratae it has a secondary transverse suture near its distal end, as shown in Sepsis (Fig. 583) and Eristalis (Fig. 588). A few genera show special modifications of the epipharynx. This is particularly true of Dolichopus (Fig. 524 and 528). In this genus the epipharynx closely resembles the hypothetical type in the presence of a distinct membrane between the labrum (1) and the epipharynx (ep). The specialization of the epipharynx consists in the bifurcation of its distal end and in the presence of a long club-shaped piece which projects from its meson dorsad into the cavity formed by the basiphar- ynx, the tormae, and the fronto-clypeus. These modifications are peculiar to species of the Dolichopodidae. The bifurcations at the distal end are of particular interest, since they have been interpreted as man- dibles by Langhoffer (1888). They are much longer in some of the genera of the family than in others. The lateral and caudal views of the epipharynx and the hypopharynx of Dolichopus show clearl.v the relation these projections have to the other parts, and justify the inter- pretation here given. The single, median, distal, lance-like portion of the hypopharynx is present in all but a few of the genera studied. The cephalic portion of the labium usually connects with the lance-like portion of the hy- popharynx just ventrad of the point of entrance of the salivary duct. 223] HEAD OF DIPTERA— PETERSON 53 In a few cases, as in Borborus (Fig. 565 and 567), the liypopliaryux is completely fused with the labium, while in others, as in Euaresta (Fig. 572), it is nearly so. In a majority of the genera the secondary separa- tion of the lance-like portion of the hypopharynx from the basipharynx corresponds with the similar separation in the epipharjTix. The shape and size of the hj^jopharj-nx also vary considerably, as can be seen in the figiires. In mouth-parts titted for sucking and piercing, the hy- popharynx is usually long and needle-like ; while in licking forms (most Calyptratae), it is greatly reduced. The salivary duct (s. d) enters the proximal portion of the lance- like part of the hypopharynx and in most cases it is carried as a duct or groove along the cephalic surface of that organ to the distal end. The course of this duct or groove is indicated by broken lines in the figures of the caudal aspect of the hypopharynx. The salivary duet before entering the hypopharjoix is enlarged and bulb-like in many species. In Tabanus (Fig. 494) the salivary bulb (s. b) is a chitinized structure continuous with the hypopharynx, while in Promachus (Fig. 517) it is chitinized, but separated from the hj-popharynx. A chitinized bulb and an enlarged membranous swelling are both present in Dolicho- pus (Fig. 528). The peculiar epipharyux and hypopharynx of Olfersia (Fig. 606) can be homologized with the more common types found thruout the order. The principal difference is in the shape and position of the basipharj-nx, the tormae, and the hyoid. The two lance-like structures embedded in the deep membranous depression about the oral cavity are the labrum-epipharynx and the lance-like part of the hypopharynx. The long, crescent-shaped piece which extends cephalad from the proxi- mal end of the labrum-epipharrax to the pear-shaped piece, is homolo- gous with the hyoid (hy), and the pear-shaped piece with which the hyoid connects is composed of the tormae (to) and the basipharynx (bpli). The exposed parts of the tormae in the membrane ventrad of the head are very small in this genus. Only rudiments of mouth-parts are found in the head of Gastrophi- lus (Fig. 490 and 492). The anterior end of the alimentary canal is a simple chitinized tube which leads to the small opening on the ventral aspect of the head. This tube undoubtedly originated from the epiphar- ynx and the hypopharynx. The mouth-parts are greatly reduced or wanting. It is possible that the small bvdb-like structures located latero-caudad of the opening are remnants of the labium. It is impos- sible to homologize the other minute modifications surrounding the mouth-opening. In the Cyrtidae, as Oncodes (Fig. 109, 486, and 487), the mouth- 54 ILLINOIS BIOLOGICAL MONOGRAPHS [224 parts show a greater reduction than in Gastrophilus, while in species of Bulonchus (Fig. 364a) they are well developed. In Oncodes a chi- tiuized ring is present in the membrane which covers the oral cavity, and a broad plate extends dorsad from its caudal margin, giving rise to a small membranous tube, the oesophagus, which has no opening to the exterior as far as could be determined. It is impossible to homolo- gize the parts within the oral cavity. The ental plate which gives rise to the oesophagus, may be homologous with the basal portion of the epipharynx and the hypopharynx. A general survey of the epipharynx and hypopharynx shows that the relationship between these parts and the head-capsule corresponds with the relationship between the mouth-parts and the head. Since the epipharynx and the hypopharynx are always connected with the labrum and the proximal part of the labium, they are projected ventrad when the labrum and labium are extruded. The interrelation of the mouth- parts and the epipharynx and hypopharynx is fixed, never changing thru- out the order, no matter what specialization may take place. The espe- cially striking feature of the epipharynx and the hpopharynx in various genera which have functional mouth-parts, is the decided similarity of the two thruout the order, as shown by the various figures. The parts undergo secondary changes in their size and shape, but in no case where the mouth-parts are functional is there an entire loss of a part, which, however, happens in many cases with the mouth-appendages. The epi- pharynx and hpopharynx of the Calyptratae in particular show a devel- opment of joints, secondary sclerites, and membranous areas, which permit a considerable amount of flexibility. SUMMAEY This investigation deals with the homology of all the sclerites of the fixed and movable parts of the head of one or more representatives of fifty-three of the fifty-nine families of the Diptera of North America as listed by Aldrich. With this large series it has been possible to make clear a number of little-understood relationships and structural modifications in the head and mouth-parts, and also to point out their homology with the corresj^onding parts and areas in insects of other orders. The six hundred and more figures show the form and structure of all the parts for each of the families studied. Modifications of the fixed and movable parts usually take the form of reduction, change of shape, loss of chitinization, or expansion of the membranous areas. The different parts have been discussed separately, and a hypothetical or typical form has been constructed for each part. 225] HEAD OF DIPTERA—PETERSON 55 Oue of the most important conclusions concerning the generalized head-capsule relates to the position of the epicranial suture. The stem of this suture along the dorso-meson represents the line of fusion of the paired sclerites of the head, while the arms of the suture ventrad of the antennal fossae enclose the unpaired sclerites of the head. This suture resembles the epicranial suture in the immature stages and the adult forms of all the generalized members of the more common orders. Two unpaired sclerites, front and clypeus, are enclosed by the fork of the epicranial suture, and in all but one or two genera form a con- tinuous area called the fronto-clypeus. The labrum is an unpaired, distinct, tongue-like structure situated ventrad of the fronto-clj-peus. It is joined to the epipharynx and the resulting structure is kno^vn as the labrum-epipharynx. The tormae are chitinized lateral pieces of the epipharynx which project cephalad and unite with the fronto-clypeus in generalized Dip- tera. They are also present in such generalized insects as the Orthop- tera. In the more specialized Diptera the tormae are interpolated be- tween the fronto-clypeus and the labrum, and in all but a few genera lose all connection with the chitinized portions of the fronto-clypeus. Their exposed surface is best seen from a cephalic view. The crescent-shaped frontal suture dorsad of the antennal fossae marks the line of invagination of the ptUinum. The origin of the ptilinum has not been determined. The vertex is the paired continuous area on the cephalic aspect of the head, and the region of the vertex ventrad and mesad of each com- pound eye is a gena. The compound eyes are usually large and located on the cephalo- lateral aspects of the head. Thej' show secondary sexual characters in a greater number of species than do anj- other of the fixed and movable parts. The three ocelli are arranged in the form of a triangle aud located on the vertex dorsad of the bifurcation of the arms of the epi- cranial suture. The occiput and postgenae are continuous areas of the caudal sur- face. The former occupies the dorsal portion and is secondarily modified about the occipital foramen to form the parocciput. The postgenae are the two areas of the ventral half, separated by a membrane in gener- alized forms and united ventrad of the occipital foramen in all the Brachycera and the Cyclorrhapha. They are also secondarily di'vided into parapostgenae along the mesal membrane. The tentorium of generalized Diptera is represented by the usual three pairs of arms and a rudimentary body. It undergoes striking modifications, aud influences to a considerable extent the detailed struc- 56 ILLINOIS BIOLOGICAL MONOGRAPHS [226 ture of the head. The relation between the invaginations of the ten- torium and the movable appendages of the mouth, wliich is so important a feature of all generalized insects, is also characteristic of the members of this order. The development of the antennae from a generalized filiform type to that found among the Cyclorrhapha can be traced on the figures. Only a few generalized Diptera have mandibles. These are only present in the females except in Simulium, in which they are well developed in both sexes. All Diptera having functional mouth-parts have maxillae. The maxillae of generalized Diptera resemble the maxillae of generalized insects except for the absence of palpifers and the fusion of the cardines and stipites with the head-capsule. The maxillae undergo considerable modification, and are reduced to a mere ental rod and a palpus in the Calyptratae. The labium is the most characteristic and specialized appendage of the mouth, and shows modifications due to reduction and membranous development. The palpigers and labial palpi are always wanting. The submentum and mentum are represented by a membranous area of the caudal surface of the head. The ligula, or the movable portion of the labium, has a basal part which usually gives rise to two large bulb-like paraglossae and to glossae situated between them. The paraglossae are specialized, and have chitiuized areas on their lateral and caudal sur- faces and pseudotracheae on their mesal surface. The parts of the epipharynx and the hypopharynx can be homolo- gized with the corresponding parts in generalized insects. There is a great similarity in the form of the epipharynx and hypopharynx of all Diptera, which is especially striking when considered in connection with the modifications that have taken place in all other parts. The various mouth-parts show striking modifications thruout the order, but all, including the epipharjrax and the hypopharjmx, retain their relative positions, even tho they may be extruded from the head- capsule for a considerable distance, as in some of the Calyptratae. The proboscis of the Cyclorrhapha is composed of the labium, maxiUae, hypopharynx, labrum-epipharynx, and tormae. The paraglossae of the labium form the large lobes, or labeUae, at its distal end. The mouth-parts of Oncodes and Gastrophilus are not functional, and are so greatly reduced that it is difficult to homologize their parts. 227] HEAD OF DIPTERA— PETERSON 57 BIBLIOGRAPHY* Becher, E. 1882. Zur Kenntnis der Mundtheile der Dipteren. Deiiksclir. k. Akad. Wissensch., Wien, math.-naturw. CI., 45:123-162; 4 pi. 1883. Zur Abwehr. Zool. Anz., 6:88-89. Berlese, A. 1909. Gli Insetti loro organizzazione, sviluppo abitudini e rapporti coll'umo., I :I54-IS9. Bl^NCHARD, E. 1850. De la Composition de la Bouche dans les Insectes de I'ordre des Dipteres. Compt. rend. Acad. Sci. Paris, 31 :424-427. Chatin, J. 1880. Sur la Constitution de I'Armature Buccale des Tabanides. Bull, de la Soc. Philom., 4, No. 8:104. Christophers, S. R. ♦1901. The Anatomy and Histology of the Adult Female Mosquito. Reports to the Malarial Committee of the Roy. Soc. (4) , 20 pp. ; 6 pi. London (Harrison and Sons). CoMSTOCH, J. H., and Kochi, C. 1902. The Skeleton of the Heads of Insects, .^m. Nat.. 36:13-45; 29 fig. Cragg, F. W. ♦1912. Studies on the Mouth-parts and Sucking Apparatus in the Blood- sucking Diptera. No. I. Philaematomyia insignis Austen. Sci. Mem. Med. Ind., No. 54; 15 pp.; 5 pi. Calcutta. DiMMOCK, G. 1881. The Anatomy of the Mouth-parts and of the Sucking Apparatus of some Diptera. Dissertation. 50 pp.; 4 pi. Boston. 1882. Anatomy of the Mouth-parts and of the Suctorial Apparatus of Culex. Psyche, 3:231-241. Frey, R. 1913. Uber die Mundteile der Mycetophiliden, Sciariden, und Cecidomyiiden. Acta Soc. Fauna Flora Fenn., Z7, No. 2. 54 pp. ; 4 pi., I fig. Gerstfeldt, G. *i853. tlber die Mundteile der saugenden Insekten. Dissertation. 121 pp.-; 4 pi. Dorpat. Gn.ES, G. M. *I905. Mouth-parts of the Biting Flies.— Preliminary Note. Jour. Trop. Med., 8:363-364. London. •An asterisk indicates publications that have not been seen by the author. 58 ILLINOIS BIOLOGICAL MONOGRAPHS [228 1906. Mouth-parts of the Biting Flies of the Genera Stomoxys and Glossina. Jour. Trop. Med., 9:99-102, 153-156, 169-173, 182-186, 198-202, 217-219, 235- 236. London. Grunberg, K. 1906. Uber bUitsaugende Musciden. Zool. Anz., 30:78-93; 15 fig. 1907. Die blutsaugenden Dipteren. Leitfaden. Jena (Fischer). Hammond, A. 1874. The Mouth of the Crane-fly. Sci. -Gossip, 1874:155-160; fig. 97-113. Hansen, H. J. 1883. Fabrica oris dipterorum : Dipterernes Mund i anatomisk og sys- tematisk Henseende. Kjobenhavn. pp. 1-220; pi. 5. Harris, W. H. 1902. The Dentition of the Diptera. Jo-jr. Quek. Micr. Club (2), 8:389- 398; pi. 19, 6 fig. Hewitt, C. G. 1907. The Structure, Development and Bionomics of the House-fly, Musca domestica, Linn. Part I. — Anatomy of the Fly. Quart. Jour. Micr. Sci., 51:395-448; pi. 22-26. HUNT^ G. 1856. The Proboscis of the Blow-fly. Quart. Jour. Micr. Sci.. 4:238-239; 2 fig. Kellogg, V. L. 1899. The Mouth Parts of the Nematocerous Diptera. Psyche, 8:303-306, 327-330, 346-348, 355-359, 363-365; II fig- 1902. The Development and Homologies of the Mouth Parts of Insects. Am. Nat., 36:683-706; 26 fig. KOLBEj H. J. 1893. Insektenkunde. 8°, 724 pp. Berlin. ICraepelin, K. 1882. tiber die Mundwerkzeuge der saugenden Insecten. Vorlaufige Mitthei- lung. Zool. Anz., 5:574-579; 3 fig. 1883. Zur Anatomic und Physiologie des Riissels von Musca. Zeitschr. wiss. Zool., 39:683-719; pi. 40-41. KULAGIN, N. 1905. Der Kopfbau bei Culex und Anopheles. Zeitschr. wiss. Zool., 83 :285- 335 ; 3 pl- Langhoffer, a. 1888. Beitrag zur Kenntnis der Mundteile der Dipteren. Dissertation. 32 pp. Jena. 1901. Mandibulae Dolichopodidarum. Verhandl. Internat. Zoologen-Con- gress, Berlin, 5:840-846; 4 fig. Leon, N. 1904. Vorlaufige Mitteilung iiber den Saugriissel der Anopheliden. Zool. Anz., 27:730-732; I fig. LowNE, B. T. 1870. The Anatomy and Physiology of the Blow-fly (Musca vomitoria, Linn.). 121 pp. ; 10 pl. 229] HEAD OF DIPTERA— PETERSON 59 1893. A Reply to some Observations on the Mouth-organs of the Diptera. Ann. Mag. Nat. Hist. (6), 11:182-184. 1890-95. The Anatomy, PhysioIog>', Morpholog>' and Development of the Blow-fly (CaUiphora erythrocephala). 778 pp. ; 52 pi. London (R. H. Porter) . MacCloskie, G. 1880. The Proboscis of the House-fly. Am. Nat., 14:153-161; 3 fig. 1884. Kraepelin's Proboscis of Musca. .\m. Nat, 18:1234-1244; 12 fig. 1888. The Poison-Apparatus of the Mosquito. Am. Nat., 22:884-888; 2 fig. Marlatt, C. L. 1896. The Mouth-parts of Insects, with particular Reference to the Diptera and Hemiptera. Abstract, Proc. A. A. A. S., 44:154-155. Meinert, F. 1880. Sur la Conformation de la Tete et sur ITnterpretation des Organes buccau.x chez les Insectes ainsi que sur la Systematique de cet Ordre. Entom. Tidskrift, i : 147-1 50. 1880a. Sur la Constitution des Organes buccaux chez les Dipteres. Entom. Tidskrift., 1:150-153. 1881. Fluernes Munddele Trophi Dipterorum. 91 pp. ; 6 pi. Kjobenhavn. 1882. Die Mundtheile der Dipteren. Zool. Anz., 5:570-574, 599-603. Menzbier, M. a. *i88o. Uber der Kopfskelett und die Mundteile der Zweifliigler. Bull. Soc. Imp. Nat Moscou, 55 :8-7i ; 2 pi. MUGGENBURG, F. H. 1892. Der Riissel der Diptera pupipara. Arch. Naturg., 58 :287-332 ; 2 pi. Packard, A. S. 1903. A Textbook of Entomology. 8°, 715 pp. New York. Patton, W. S., and Cragg, F. W. 1913. A Textbook of Medical Entomologj-. London, Madras, and Calcutta (Christian Literature Societ>' for India). Pazos y Caballera, F. H. 1903. Del exterior e interior del Mosquito. Apuntes sobre la .Anatomic y Morfologia. Rev. Trop. Med., 4:209-218; 4 pi. Peterson, Alvah 1915. Morphological Studies on the Head and Mouth-parts of the Thy- sanoptera. Ann. Ent. Soc. Amer., 8 :20-67 ; 7 pi. Smith, J. B. 1890. A Contribution to a Knowledge of the Mouth Parts of the Diptera. Trans. Am. Ent Soc, 17:319-339; 22 fig. Stephens, J. W. W., and Newstead, R. *I90". The Anatomy of the Proboscis of Biting Flies. Ann. Trop. Med., 1:171-198; 8 pi. Liverpool. Steinheil, a. P. 1896. Trophi of Tabanus. Studies from the Lab. Zool., Dipt., Imp. Univ. Warsaw, 1896:247-250; 2 fig. 60 ILLINOIS BIOLOGICAL MONOGRAPHS [230 Suffolk, W. T. 1869. On the Proboscis of the Blow-fly. Monthly Micr. Jour., 1:331-342; 4 pl- Wateehouse, C. O. ♦1893. Some Observations on the Mouth-organs of Diptera. Ann. Mag. Nat Hist. (6), 11:45-46. Wesch^ W. 1902. Undescribed Palpi on the Proboscis of some Dipterous Flies, with Remarks on the Mouth-parts in several Families. Jour. Roy. Micr. Soc, 1902:412-416; 2 pl. ♦1903. The Mouth-parts of the Tsetse-fly. Knowledge, 26:116-117; i fig. 1904. The Labial and Maxillary Palpi in Diptera. Trans. Linn. Soc. Lon- don, 9:219-230. 1906. The Genitalia of both Sexes in Diptera and their Relation to the Armature of the Mouth. Trans. Linn. Soc. London, 9:339-386; 8 pl. 1908. The Proboscis of the Blow-fly, Calliphora erythrocephala. A Study in Evolution. Jour. Quek. Micr. Club (2), 10:283-295; 2 pl. 1909. The Mouth-parts of the Nemocera and their Relations to the other Families in Diptera. Corrections and Additions to the Paper published in 1904. Jour. Roy. Micr. Soc, 1909:1-16; pl. 1-4 1912. The Phylogeny of the Nemocera, with Notes on the Leg Bristles, Hairs and certain Mouth Glands of Diptera. Biol. Bull., 23 :2SO-270 ; i pl. 231] HEAD OF DIPTERA— PETERSON 61 EXPLANATION OF PLATES ABBREVIATIONS USED a.a Anterior arms of the tentorium i.a.d Invagination of the anterior and a.e.s Arms of the epicranial suture dorsal arms of the tentorium a.f Antennal fossa i.d Invagination of the dorsal arm al.c Alimentary canal of the tentorium ant Antenna i.p Invagination of the posterior ar Arista arm of the tentorium a.s Antennal sclerite k Kappa (sclerite) bph Basipharynx 1 Labrum bpr Basiproboscis la Lacinia b.t Body of the tentorium le Labella c Clypeus l.ep Labrum epipharynx ca Cardo Ig Ligula c.e Compound eye li Labium ch Chitinized m Membrane ch.th Chitinized thickening md Mandible c.l.s Clypeo-labral suture me Mentum cu Cornu mpr Mediproboscis d.a Dorsal arms of the tentorium m.ps Main pseudotracheae de Depression mx Maxilla dpr Distiproboscis nix.pl Maxillary palpus ep Epipharynx n.s Neck sclerite e.s Epicranial suture oc Ocellus f Furca, also f-i, f-2, and f-3 oca Ocellar area fa Facet occ Occiput fl Flagellum oe Oesophagus fr Front oe.p Oesophageal pump fr.c Fronto-clypeus o.f Occipital foramen fr.s Frontal suture 0.1 Oral lobe g Galea o.s Ocular sclerite ge Gena p.a Posterior arms of the tentorium gl Glossa pd Pedicel h Hook pgl Paraglossa hp Hypopharynx po Postgena hy Hyoid pocc Parocciput i.a Invagination of the anterior arm ppo Parapostgena of the tentorium pr Proboscis 62 ILLINOIS BIOLOGICAL MONOGRAPHS [232 ps Pseudotrachea so Sense organ ps.th Pseudotracheal thickening s.s Secondary suture Pt Ptilinum St Stipes, st-i and st-2 ectal part, r.d.a Rudimentary dorsal arms of the st-e ental part tentorium su Submentum r.p.a Rudimentary posterior arms of su.me Submentum and mentum the tentorium t Tentorium s Suture tee Teeth-like structures s.b Salivary bulb th Thickening sc Scape the Theca s.d Salivary duct to Torma or tormae s.e.s Stem of the epicranial suture t.th Tentorial thickening si Sigma (sclerite) V Vertex 233] HEAD OF DIPTERA— PETERSON 63 PLATE I 64 ILLINOIS BIOLOGICAL MOSOGRAPHS [234 EXPLANATION OF PLATE Cephalic Aspect of the Head and Mouth-parts Fig. I. Hj'potlietical head. Fig. 2. Simulium vcnustum , female. Fig. 3- Simulium johannseni, male. Fig. 4- Bibiocephala elcgantula, male. Fig. s. Bibiocephala elegantula, female. Fig. 6. Rhabdophaga strobiloides. Fig. 7- Mycctobia divcrgcns. Fig. 8. Psychoda albipennis. Fig. 9- Rhyphus punctatus. Fig. 10. Psorophora ciliata, female. Fig. II. Mycctophila punctata, female. Fig. 12, Chironomus ferrugineovittatus, female. Fig. 13- Bibio femorattis, male. Fig. 14- Bibio femoratus, female. Fig. IS- Ptychoptcra rufocincta. Fig. 1 6. Trichoccra bimacula. Fig. 17- Sciara varians. Fig. i8. Tipula bicornis. < ILUXOIS BIOLOGICAL MONOGRAPHS VOLUME "^'" @ PETERSOX HEAD AND MOUTH PARTS OF DIPTERA PLATE I I 235] HEAD OF DIPTERA—FETERSOX 65 PLATE II 66 ILLIXOIS BIOLOGICAL MONOGRAPHS [236 EXPLANATION OF PLATE Cephalic aspect of the Head Fig. 19. Di.va clavata. Fig. 20. Tabanus gigantcus, female. Fig. 21. Tabanus gigantcus, male. Fig. 22. Proiuachus vertebratus. Fig. 23. Eristalis tenax, female. Fig. 24. Eristalis tenax, dorsal end of the tormae. Fig. 25. Eristalis tenax, male. Fig. 26. Psorophora ciliata, male. Fig. 27. Stratiomyia apicula, male. Fig. 28. Stratiomyia apicula, female. Fig. 29. Exoprosopa fasciata. Fig. 30. Mydas clavatus. Fig. 31. Aphiochacta agarici. Fig. 32. Platypeza velutina. •Fig. 2i. Psilocephala haciuorrhoidalis, male. Fig. 34. Lcptis vertebrata, female. Fig. 35. Lcptis vertebrata, male. Fig. 36. Psilocephala hacmorrhoidalis, female. Fig. 37. Lonchoptcra lutea, female. Fig. 38. PipuncuUts ciiigulatus, female. IIJJXOIS BIOLOGICAL MOXOCRAPHS rOLUME s PETERSON HEAD AND MOUTH PARTS OF DIPTERA PLATE H * 237] HEAD OF DIPTERA— PETERSON 67 PLATE III 68 ILLINOIS BIOLOGICAL MONOGRAPHS [238 EXPLANATION OF PLATE Cephalic Aspect of the Head Fig. 39. Pipunculus cingulatus, male. Fig. 40. Empis clausa, female. Fig. 41. Scenopinus fenestraUs, male. Fig. 42. Scenopinus fencstralis, female. Fig. 43. Dolichopus bifractus. Fig. 44. Calobata univitta. Fig. 45. Drosophila ampelophila. Fig. 46. Sepsis violacea. Fig. 47. Desmometopa latipes. Fig. 48. Oecothea fcnestralis. Fig. 49. Heteroneura flaviseta. Fig. 50. Chyromya concolor. , Fig. 51. Chloropisca glabra. Fig. 52. Sphyraccphala hrcvicornis. Fig. 53. Oncodes costatus. Fig. 54. Gastrophilus equi. Fig. 55. Tetanocera pluntosa. Fig. 56. Ochthera mantis. Fig. 57. Oljersia ardeae. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 PETERSON HEAD AND MOUTH PARTS OF DIPTERA PLATE HI 239] HEAD OF DIPTERA — PETERSON 69 PLATE IV ILLINOIS BIOLOGICAL MOXOGRAPHS [240 EXPLANATION OF PLATE Fig 58 Fig 59 Fig 60 Fig 61 Fig 62 Fig 63 Fig 64 Fig 65 Fig 66 Fig 67 Fig 68 Fig 69 Fig 70 -Fig 71 Fig 72 Cephalic Aspect of the Head Coclopa vanduzeii. Loxocera pectoralis. Saproiiiy::a vulgaris. Euaresta aequalis. Scatophaga furcata. Borborus equiims. Cbrysoviysa dcmandata. Thelaira leucosona. Sarcophaga haeiiwrrhoidalis. Conops hrachyrhynchns. Archytas aiialis. Hydrotaea dentipcs, female. Hydrotaea dentipes, male. Musca domestica, female. Miisca domestica, male. ILLIXOIS BIOLOGICAL MONOGRAPHS VOLUME s PETERSOX HEAD AND MOUTH PARTS OF DIPTEKA PLAll-; IV I 241] HEAD OF DIPTERA— PETERSON PLATE V ILLINOIS BIOLOGICAL MONOGRAPHS [242 EXPLANATION OF PLATE Fig 73 Fig VFig 74 75 Fig 76 Fig 77 Fig 78 Fig 79 Fig 8o Fig 8i Fig 82 Fig 83 Fig 84 Fig 8S Fig 86 Fig 87 Fig 88 Fig 89 Fig 90 Caudal Aspect of the Head Hypothetical head. Tabanus giganteus, female. Tnbanus giganteus, male. Bibiocephala clegantula, male. Simuiium venustum, female. Trichocera bimacula. Dixa clavata. Rhyphus functatus. Sciara varians. Psyclioda albipennis. Bibiocephala elegantula, female. Promachus vertebratus. Bittaconwrpha clavipes. Rhabdophaga strobiloides. Mycetophila punctata. ChiroHOmus ferrugineovittatus. Chirono)nus ferrugineovittatus, dorsal aspect. Mycetobia divergens. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 PETERStJX IIKAIJ AM) .MOUIll PARTS OF DIPTERA PLATE V 243] HEAD OF DIPTERA—PETERSOX 73 PLATE VI 74 ILLINOIS BIOLOGICAL MONOGRAPHS [244 Fig. 91 Fig. 92 Fig. 93 Fig. 94 Fig. 95 Fig. 96 Fig. 97 Fig. 98 Fig. 99 Fig. 100 Fig. lOI Fig. 102 Fig. 103 Fig. 104 Fig. 105 Fig. 106 Fig. 107 Fig. 108 Fig. 109 Fig. 110 Fig. III Fig. 112 EXPLANATION OF PLATE Caud.\l Aspect of the Head Bibio femoratus, male. Bibio femoratus, female. Limnobia immatura. Sphyracephala brcvicornis. , Tiptila bicornis. Psorophora ciliata, female. Enipis clausa, female. Exoprosopa fasciata. Mydas clavatus. Psilocephala haeiuorrhoidalis, female. Ochthera mantis. Lonchoptera Iiitea, female. Leptis vertcbrata, male. Stratiomyia apicula. male. Oncodes costatus. Pipuncuhis cingiilatus, female. Scenopinus fenestralis. Dolichopus sp. Oncodes costatus. ventral aspect. Platypeca vclutina. Aphiochaeta agarici. Dolichopus bifractits, lateral margins incomplete. ILLIXOIS niOLOCICAI. MOXOGR'.irilS ■ GLUME 3 PE'lT-RSOX lll'.AI) AM) AKil'l II I'AKTS ol' DIl'TCKA i'l.AI I-: \'l 245] HEAD OF DIPTERA— PETERSON 75 PLATE VII 76 ILLINOIS BIOLOGICAL MOXOCRAPHS [246 EXPLANATION OF PLATE Caudal Aspect of the Head Fig. IS- Eristalis tenax, female. Fig. M- Calobata univitta. Fig. IS. Sapromysa vulgaris. Fig. i6. Lispa nasoni, margin incomplete. Fig. 17. Conops brachyrhynclius. Fig. 18. Sepsis violacea. Fig. 19- Tetanocera plumosa. Fig. 20. Myiospila meditabunda, margin incomplete Fig. 21. Coelopa vandiiscii. Fig. 22. Chiromya concolor. Fig. 23- Loxocera pectoralis. Fig. 24. Archytas analis. Fig. 25. Drosophila ampclophila. Fig. 26. Heteroneura flaviseta. Fig. 27. Hydro taea dentipes. Fig. 28. Thelaira leucozona. Fig. 29- Desmomctopa latipes. Fig. 30. Sarcophaga hacmorrhoidalis. Fig. 31- Euarcsta aeqtialis. Fig. 32. Chloropisca glabra. Fig. 33- Musca domestica, female. Fig. 34- Chrysoniyza demandata. Fig. 35- Scatophaga furcata. Fig. 36. Borborus equinus. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 PETHRSOX 11I-.\I) AM) MOITH PARTS OF DIPTF.RA PLA'l I'. \'II 247] HEAD OF DIPTERA— PETERSON 77 PLATE VIII 78 ILLIXOIS BIOLOGICAL MONOGRAPHS [248 EXPLANATION OF PLATE Caudal and Lateral Aspects of the Head and the Tentorium Oecothea fenestralis, caudal aspect. Gasirophilus cqui, caudal aspect. Olfersia ardcac. caudal aspect. Hypothetical head, lateral aspect. Hypothetical tentorium, lateral aspect. Tabunus giganleus, female, lateral aspect. Tabanus giganteus, lateral aspect of the tentorium. Siitniliuiii vcnustum, female, lateral aspect. Leptis vcrtebrata, male, lateral aspect. Mydas clavatus, lateral aspect. Promachus vertebratus, lateral aspect. Promachus vertebratus, lateral aspect of the tentorium. Scenopinus fenestralis, female, lateral aspect. Sciara variaus, lateral aspect. Pipunculus cingulatns, lateral aspect. Chironoinus fcrrugincovittatus, lateral aspect. Bibio feiiioratus, female, lateral aspect. Bibio fciijoratus, male, lateral aspect. Fig. 37- Fig. 38. Fig. 39 Fig. 40. Fig. 41 Fig. 42 Fig. 43 Fig. 44 Fig. 45 Fig. 46 Fig. 47 Fig. 48 Fig. 49 Fig. SO Fig. 51 Fig. 52 Fig. 53 Fig. 54 I ILLISOIS BIOLOGICAL MONOGRAPHS VOLUME 3 'F/rERSON HEAD AND MOUTH PARTS OF DIPTERA PLATE VIII -249] HEAD OF DIPTERA — PETERSOX 79 PLATE IX 80 ILLIXOIS BIOLOGICAL MONOGRAPHS [250 EXPLANATION OF PLATE ?AL Aspect of the Head showing the Tentorium Fig. 155 Bibioccphala elegantula, female. Fig. 156 Bibiocephala elegantula, male. Fig. 157 Rhyphus punctatiis. Fig. 158 Trkhocera biiiiacula. Fig. 159 Psorophora ciliata, female. Fig. 160 Stratioinyia apiciila, male. Fig. 161 Mycetobia divergcns. Fig. 162 Exoprosopa fascia ta, eye removed. Fig. 163 Dixa clavata. Fig. 164 Eiiipis clav.sa, female. ^ig. 165 Platypcza velutina. -Mg. 166 Psychoda albipennis. Fig. 167 Eristalis tcnax, female, eye removed. Fig. 168 Dolichopus bifractus, eye removed. Fig. 169 Loxocera pectoralis. Fig. 170 Rhabdophaga strobiloides. Fig. i/l Saprowyza vulgaris. Fig. 172 Drosophila ampclnphila. Fig. 173 Psilocephala haemorrlwidalis, female. Fig. 174 Aphiochaeta agarici. Fig. 175 Euaresta acqualis. Fig. 176 Heteroneura flaviseta. Fig. 177 Lonchoptera lutea. Fig. 178 Tipula bicornis. Fig. 179 Chyromya concolor. ILLINOIS BIOLOGICAL MONOGRAPHS VOLUME 3 PMIHRSOX llF.Al) AXD AIOL'Jll PARTS OF DIPTERA PLATE IX 251] HEAD OF DIPTERA— PETERSON PLATE X 82 ILLIXOIS BIOLOGICAL MOKOCRAPHS [2S2 EXPLANATION OF PLATE Lateral Aspect of the Head showing the Tentorium Fig. l8o. I'etanocera pluinosa. Fig. i8i. Chrysoiuy^a deiitandala. Fig. 182. Coelopa vanduseii. Fig. 183. Calobata univitta. Fig. 184. Sepsis violacea. Fig. 185. DcsiiwiJictopa latipes. Fig. 186. Conops brachyrhynchus. Fig. 187. Ochtherti iiianlis. Fig. 188. Borbnrus cquinus. Fig. 189. Chloropisca glabra. Fig. iQph. Hypothetical antenna. Fig. 199. Dixa clavata. Fig. 200. Trichocera bimacula. Fig. 201. Rhabdophaga strobiloides. Fig. 202. Psychoda aWipennis. Fig. 203. Bibiocephala elcgantula. Fig. 204. Simuliuiu venustum. Fig. 205. Sciara zarians. Fig. 190 Sphyraccphala brcvicornis. Fig. 191 Sarcophaga haenwrrhoidalis. Fig. 192 Occothea fciiestralis. Fig. 193 Scatophaga furcata. Fig. 194 Musca domestica. Fig. 195 Hydrotaca dentipes. Fig. 196 Tliclaira leucocona. Fig. 197 Arcbytas analis. Fig. 198 Olfcrsia ardeae. INAE Fig. 206 Chironoinus fcrrugiiteovitta tiis, female. Fig. 207 Chironoinus ferrugin eovitta tus, male. Fig. 208 Bihio fei'ioratus, female. Fig. 209 Rliyphus punctatus. Fig. 210 Psorophora ciliata, female. Fig. 211 Psorophora ciliata, male. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME s i'I-:i KKSOX HEAD AND .MOUTH PARTS OF DIPTERA PLATE X 253] HEAD OF DIPTERA — PETERSOX 83 PLATE XI 84 ILLINOIS BIOLOGICAL MONOGRAPHS [254 EXPLANATION OF PLATE Antennae Fig. 212. Mydas clavatnis. Fig. 231. Fig. 213. Stratioinyia apicula. Fig. 232. Fig. 214. Tabanus giganteus. Fig. 233. Fig. 215. Empis clausa. Fig. 234. Fig. 216. Exoprosopa fasciata. Fig. 235. Fig. 217. Proinachus vcrtehratus. Fig. 236. Fig. 218. Leptis vertebra ta. Fig. 237. Fig. 219. Scenopinus fcnestralis. Fig. 238. Fig. 220. Oncodes costatus. Fig. 239. Fig. 221. Conops brachyrhynchus. Fig. 240. Fig. 222. Platypeca velutina. Fig. 241. Fig. 223. Lonchoptera lutea. — ^Fig. 242. Fig. 224. Aphiochaeta agarici. Fig. 243. Fig. 225. Tetanocera plumosa. Fig. 244. Fig. 226. Dolichopus bifratitus. Fig. 245. Fig. 227. Oecothea fcnestralis. Fig. 246. Fig. 228. Desinouiciopa latipes. Fig. 247. Fig. 229. Heteronatra flaviscta. Fig. 248, Fig. 230. Thelaira levccozona. Fig. 249. Mandibles Fig. 250. Siiiiuliuiii venustuni, female. Fig. 254. Fig. 251. Psorophora cUiata, female. Fig. 255. Fig. 252. Simuliuin johannseni, male. Fig. 256. Fig. 253. Culicoidcs sanguisugus, female. Borborus equinus. Eristalis tena.v. Chyromya coiicolor. Sepsis violacea. Loxocera pectoralis. Calobata univitta. Ochthera mantis. Drosophila ainpelophila. Gastrophilus cqui. Euaresta aequalis. Hydrotaea dcntipes. Musca doinestica. Pipunculus cingulatus. Sarcophaga haemorrhoidalis. Chrysomysa demandata. Scatophaga furcota. Archytas analis. Sapromyca vulgaris. Olfersia ardeae. Di.va modesta, female. Tabanus giganteus, female. Bibiocephala elegantula, female. ILUNOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 ---tf^jB^ 226 Dc„,.of„. ^- \; ; 245 227 (v,c,t„ (W5 / 244 PETERSON" HEAD AND MOUTH PARTS OF DIPTERA PLATE XI 255] HEAD OF DIPTERA— PETERSON 85 PLATE XII 86 ILLISOIS BIOLOGICAL MONOGRAPHS [2S& EXPLAN'ATION OF PLATE Mandible and Maxillae Fig. 256I1. Hypothetical mandible. Fig. 257. Hypothetical maxillae. Siiniiliuin venustuiii, female, cephalic aspect. Tabaitus giganteus, female, caudal aspect. Trichoccra bimanda, caudal aspect. Rliyphus punctatus, caudal aspect. Di.v(i clavata, caudal aspect. Psychoda albipennis, caudal aspect Bibio fenwratus, caudal aspect. CuUcoidc's satiguisugus, female, caudal aspect. Psorophora ciliata, female and male, caudal aspect. Sciara varians, caudal aspect. Rhabdophaga strobiloidcs, caudal aspect. Bibiocephala eleganiula, female, caudal aspect. Chironomus ferriigiveovittatus, cephalic aspect. Mydas clavatus, lateral aspect. Platypeza veluiina, lateral a^spect. Stratioiiiyia apicuhi, cephalic aspect. Einpis clausa. lateral aspect. Lcptis vertebrata, caudal aspect. Promachus vcrtebratus, caudal aspect. Tipula bicoriiis, portion of caudal aspect. Aphiochaeta agarici, lateral aspect. Pipunculus cingulatus, lateral aspect. Lonchoptera lutea. Psiloccphala haenwrrhoidalis, cephalic aspect. Scenopinits fenestralis. Tabaiius giganteus, male, caudal aspect. Dolichopus bifractus. Fig. 258. Fig. 259. Fig. 260. Fig. 261. Fig. 262. Fig. 263. Fig. 264. Fig. 265. Fig. 266. Fig- 267. Fig. 268. Fig. 269. Fig. 270. Fig. 271. Fig. 272. Fig. 273. Fig. 274. Fig. 275- Fig. 276. Fig. 277- Fig. 278. Fig. 279- Fig. 280. Fig. 281. Fig. 282. Fig. 283. Fig. 284. ILLIXOIS BIOLOGICAL MONOGRAPHS VOLUME 3 279 Pipuna>l PETERSON HEAD AND MOUTH PARTS OF DIPTERA PLATE XH 257] HEAD OF DIPTERA— PETERSON 87 PLATE XIII ILLINOIS BIOLOGICAL MONOGRAPHS [258 EXPLANATION OF PLATE Maxillae Fig. 284a. Eidonchus tristis. Fig. 285. E.xoprosopa fasciata. Fig. 286. Eristalis tenax. Fig. 287. Sepsis violacca. Fig. 288. Coelopa vandu:eii. Fig. 289. Saproviysa vulgaris. Fig. 290. Oecotliea fenestralis. Fig. 291. Drosophila ainpeiophila. Fig. 292. Euarcsta aegualis. Fig. 293. Sphyracephala brevicornis. Fig. 294. Borborus equinus. Fig. 295. Chrysomyza deniandata. Fig. 296. Calobata univitta. Fig. 297. Ochthera mantis. Fig. 298. Hetcroneura flaviseta. Fig. 299. Chyromya concolor. Fig. 300. Loxocera pectoralis. Fig. 301. Thelaira leucocona. Fig. 302. Tctanoccra pluiiwsa. Fig- 303- Dcsinometopa latipes. Fig. 304. Musca domes tica. Fig. 305. Conops brachyrhynchus. Fig. 306. Chloropisca glabra. Fig. 307. Scatophaga furcata. Fig. 308. Hydrotaea dcntipes. Fig. 309. Archytas analis. Fig. 310. Sarcophaga hacniorrhoidalis. ILLINOIS BIOLOGICAL MONOGRAPHS VOLUME 3 305«Kp 306 /"/ S"4°P1'«* ,^, Conop, Chloropisca '^ 3U ' 3U» Hydrou.:. 3O9 A„by,„ 310s.,cophi«. PF.TF.RSOX HEAD AND MOUTH PARTS OV DIPTRRA PLATF. XIII 259] HEAD OF DIPTERA— PETERSON PLATE XIV 90 ILLIXOIS BIOLOGICAL MOXOGRAPHS [26a EXPLANATION OF PLATE Lateral Aspect of the Mouth-parts or Proboscis Fig. 311 Fig. 312 Fig. 313 Fig. 314. Fig. 31S Fig. 316 Fig. 31 Fig. 31 Fig. 319. Fig. 320. Fig. 321 Fig. 322 Fig. 323 Fig. 324 Fig. 325 Fig. 326. Fig. 327 Fig. 328. Fig. 329. Fig. 330. Fig. 331 Fig. 332 Fig. 333 Trichocera biinacula. Chironoimis ferrugineovittatus. Rhabdophaga strobiloides. Sciara varians. Bibio feiiwratus. SimuUum venustuiii, female. Tabanus giganteus, female. Psychoda albipennis. Mydas clavatus. Lonchoptera lutea. Rhyphus punctatus. Promachus vertebratus. Leptis vertehrata. Psiloccphala haeiiiorrlwidalis. Scenopinus fencstralis. Platypeza velutina. PipuHculus cingulattis. Eristalis tenax. Sapromyza vulgaris. Desmoiuetopa latipes. Stratiomyxa apicula. Oecothea fcnestralis. Chyroiuya concolor. ILl.IXOIS BIOLOGICAL MONOGRAPHS VOLUME 3 327 p,p„„c„iu, 328 E„ PETERSON HEAD AND MOUTH PARTS OF DIPTERA PLATE XIV 261] HEAD OF DIPTERA— PETERSON 91 PLATE XV 92 ILLINOIS BIOLOGICAL MONOGRAPHS [262; EXPLANATION OF PLATE Lateral Aspect of the Proboscis Fig- 334- Sepsis violacea. Fig. 335. Aphiochaeta agarici. Fig. 336. Ochthera mantis. Fig. Z2~. Coelopa vanduzeii. Fig. 338. Sphyracephala brevicornis. Fig. 339. Loxocera pectoralis. Fig. 340. H ctcroneura flaviseta. Fig. 341. Chrysoiiiyza deinandata. Fig. 342. Borborus eguinus. Fig. 343. Drosophila ainpelophila. Fig. 344. Tetanoccra plumosa. Fig. 345. -Chloropisca glabra. Fig. 346. Thelaira leucozona. Fig. 347. Euaresta aequalis. Fig. 348. Calobata univitta. Fig. 34Q. Hydrotaca dcntipcs. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME s PETERSON HEAD AXD MOUTH PARTS OF DIPTERA PLAri'. X\' 263] HEAD OF DIPTERA— PETERSON 9i PLATE XVI 94 ILLIXOIS BIOLOGICAL MOXOCRAPHS [264 EXPLANATION OF PLATE Fig. 350. yFig. 351 Fig. 352. Fig. 353 Fig. 354. Fig. 355 Fig. 356. Fig. 357 Fig. 3S8. Fig. 359 Fig. 360, Figr 361 Fig. 362 Fig. 363, Fig. 364 Fig. 364a Fig. 365 Fig. 366. Fig. 367, Fig. 368, Fig. 369, Fig. 3-0, Mouth-parts Siircophaga haeiitorrhoidalis, lateral aspect. Miisca domcstica, lateral aspect. Einpis clausa, lateral aspect. Archytas analis, lateral aspect. Stoiiio.rys calcitrans, lateral aspect. Sipliona genicitlata, lateral aspect. Conops bracbyrhynchus, lateral aspect. Scatophaga furcata, lateral aspect. Olfersia ardeae, lateral aspect. Stylogaster biannulata, caudal aspect. Sciara varians, maxillae and labium, cephalic aspect. E.voprosopa fasciata, lateral aspect. Hypothetical and typical labium, mesal aspect. Hypothetical mouth-parts, lateral aspect. Bibio fenioratus, maxillae and labium, cephalic aspect. Eulonchus tristis, head and mouth-parts, lateral aspect. Trichocera biiuacula, maxillae and labium, cephalic aspect. Sbnulium venustuni, maxillae and labium, cephalic aspect. Rhabdophago strobiloidcs, maxillae and labium, caudal aspect, Lcia oblcctabilis, maxillae and labium, cephalic aspect. Lcptis vcrtcbrata, mesal aspect of glossa. Leptis vcrtcbrata, maxillae and labium, caudal aspect. ILLIXOIS BIOLOGICAL MOKOGRAPHS VOLUME 3 PETERSON HEAD AND MOUTH PARTS OF DIPTERA PLATE XVI 265] HEAD OF DIPTERA— PETERSON 95 PLATE XVII 96 ILUXOIS lilOLOCICAL MOXOCRAPHS [266 Fig. 3/1. Fig- 3/2. Fig. 273- Fig. 374. Fig. 375. Fig. 376. Fig. 377- Fig. 378. Fig. 379. EXPLANATION OF PLATE . Maxillae and Labium Chirononnis fcrriiyineovittatus, cephalic aspect. Psyclioda albi/'cnins, cephalic aspect. Psoro/'liora ciliata, female, portions of mandibles, maxillae, labium, ten- torium, and head-capsule. Rhyphus puiictatus, cephalic aspect. Di.va davata, cephalic aspect. Proinachus vertcbratus, caudal aspect. Proiiiachus vcrtehratus, labium, ceplialic aspect. Proinachus vcrtehratus, cross-section of labium, see figure 377. Proinachus vcrtehratus, distal end of labium, cephalic aspect. Fig. 380. Psorophora ciliata, distal end of labium, caudal aspect. Fig. 381. Psorophora ciliata, distal end of labium, cephalic aspect. Fig. 382. Geranoinyia canadensis, cephalic aspect. Fig. 383. Tipula bicornis, distal end of labium, mesal aspect. Fig. 384. Tipula bicornis, caudal aspect of labium. Fig. 385. Hclobia punctipennis, caudal aspect. Fig. 386. Limnobia innnatura, caudal aspect. Fig. 387. Dixa clavala. caudal aspect of labium. Fig. 388. Tipula bicornis, sclerites about distal end of theca of labium. Fig. 389. Bittacomorpha clavipcs, distal end of labium, mesal aspect. Fig. 390. Tabanus gigantcus, mesal aspect of labium. Fig. 391. Tabanus giganteus, caudal aspect of labium. Fig. 392. Tabanus giganteus, cephalic aspect of labium. Fig. 393. Aphiochacta agarici, caudal aspect. Fig. 394. Aphiochacta agarici, distal end of labium, mesal aspect. II.LIXOIS BIOLOGICAL MONOGRAPHS VOLl'ME 3 PETERSON HEAD AND MOUTH PARTS OF DIPTERA PLATE XVH 267] HEAD OF DIPTERA—PETERSOX 97 PLATE XVIII ILLIXaS BIOLOGICAL MOXOGRAPHS [26S EXPLANATION OF PLATE Labium Fig. 395. Stratioiiiyia apicula, caudal aspect of proboscis. Fig. 396. Stratioiiiyia apicula, mesal aspect. Fig. 397. Mydas clavatus, caudal aspect. Fig. 398. Mydas clavatus, cephalic aspect. Fig. 399. Bibioccphala elcgantiila, cephalic aspect. Fig. 400. Scenopinus fencstralis, mesal aspect. Fig. 401. Scenopinus fenestralis, caudal aspect. Fig. 402. Psilocepltala hacmorrhoidalis, caudal aspect. Fig. 403. Psilocepltala hacmorrhoidalis, mesal aspect. Fig. 404. Desiiioiuetopa latipcs, caudal aspect. Fig. 405. Desiiioiuetopa latipcs, cephalic aspect. Fig. 406. Lonchoptera lutea, caudal aspect. Fig. 407. Lonchoptera lutea, cephalic aspect. Fig. 408. Lonchoptera lutea, mesal aspect. Fig. 4,09. Saproinyso vulgaris, caudal aspect. Fig. 410. Saproinyca vulgaris, mesal aspect. Fig. 411. Chyroiiiya concolor, caudal aspect. Fig. 412. Chyroniya concolor, mesal aspect. Fig. 413. Euaresta aequalis, caudal aspect. Fig. 414. Euaresta aequalis, mesal aspect. Fig. 415. Platypeca velutina, mesal aspect. Fig. 416. Platypesa velutina, caudal aspect. Fig. 417. Conops brachyrhynchus. distal end. caudal aspect. Fig. 418. Conops brachyrhynchus, distal end, lateral aspect. Fig. 419. Conops brachyrhynchus, distal end, cephalic aspect. Fig. 420. Conops brachyrhynchus, caudal aspect. Fig. 421. Eiiipis clausa, caudal aspect. Fig. 422. Eiiipis clausa, portion of cephalic aspect. Fig. 423. Eiiipis clausa, cephalic aspect. Fig. 424. Rhainphoniyia glabra, caudal aspect. Fig. 425. Rhainphoniyia glabra, mesal aspect. Fig. 425a. Eulonchus tristis, cephalic aspect. Fig. 425b. Eulonchus tristis, distal end, mesal aspect. Fig, 426. Exoprosopa fasciata, distal end, caudal aspect. Fig. 427. Exoprosopa fasciata, cephalic aspect. Fig. 428. Exoprosopa fasciata, distal end, mesal aspect. Fig. 429. Exoprosopa fasciata, caudal aspect. ILUXOIS BIOLOGICAL MONOGRAPHS rOLCME s Einpi* Rh«mphomy a 483 424 426 427 428 4ie PETERSON HEAD AND .MOUTH PARTS OF DIPTERA PLATE W 269] HEAD OF DIPTERA— PETERSON 99 PLATE XIX 100 ILLINOIS BIOLOGICAL MONOGRAPHS [270 EXPLANATION OF PLATE Labium Fig. 430. Chloropisca glabra, caudal aspect. Fig. 431. Chlorofisca glabra, cephalic aspect. Fig. 432. Dolichopus bifractus, mesal aspect. Fig. 433. Dolichopus bifractus, caudal aspect. Fig. 434. Dolichopus bifractus. lateral aspect. F'g- 435- Pipunculus cingulatus, caudal aspect. Fig. 436. Pipunculus cingulatus, cephalic aspect. Fig- 437- Borborus cquinus, caudal aspect. Fig. 438. Borborus equinus, mesal aspect. Fig. 439. Sepsis violacea, caudal aspect. Fig. 440. Sepsis violacea, mesal aspect. Fig. 441. Eristalis tenax, mesa! aspect. Fig. 442. Eristalis tenax, caudal view. Fig. 443. Eristalis tenax, distal end of theca, caudal aspect. Fig. 444. Ochthera mantis, caudal aspect. Fig. 445. Ochthera mantis, mesal aspect. Fig. 446. Calobata univitta, mesal aspect. Fig. 447. Calobata univitta, caudal aspect. Fig. 448. Coelopa vanduzcii, caudal aspect. Fig. 449. Coelopa vandu:cii, mesal aspect. Fig. 450. Sphyracephala brcvicornis, caudal aspect. Fig. 4SI. Sphyracephala brevicornis, mesal aspect. Fig. 452. Oecothea fenestralis, caudal aspect. Fig. 453. Oecothea fenestralis, mesal aspect. Fig. 454. Drosophila ampclophila, caudal aspect. Fig. 455. Drosophila ampclophila, mesal aspect. Fig. 456. Chrysomyca demandata, mesal aspect. Fig. 457. Chrysomyca demandata, caudal aspect. Fig. 458. Siphona geniculata, distal end, cephalic aspect. ILLISOIS BIOLOGICAL MOXOGRAPHS VOLVME .? PETERSON' HEAD AND MULTll PARTS OF DIPTERA PLATE XIX 271] HEAD OF DIPTERA — PETERSON 101 PLATE XX 102 ILLINOIS BIOLOGICAL MONOGRAPHS [272 EXPLANATION OF PLATE Labium and other Parts Fig. 459. Heteroneura ftaviseta, caudal aspect. Fig. 460. Heteroneura flaviseta, mesal aspect. Fig. 461. Loxocera pectoralis, caudal aspect. Fig. 462. Loxocera pectoralis, mesal aspect. Fig. 463. Tetanocera plumosa, caudal aspect. Fig. 464. Tetanocera plumosa, mesal aspect. »^ Fig. 465. Musca domestica, dorsal aspect of glossae. Fig. 466. Musca domestica, caudal aspect. Fig. 467. Musca domestica, mesal aspect. Fig. 468. Archytas analis, caudal aspect. Fig. 469. Archytas analis, mesal aspect. Fig. 470. Scatophaga furcata, caudal aspect of mediproboscis. Fig. 471. Scatophaga furcata, ventral aspect of distiproboscis. Fig. 472. Scatophaga furcata, mesal aspect. Fig. 473. Thelaira leucosona, caudal aspect. Fig. 474. Thelaira leucocona, mesal aspect. Fig. 475. Hydrotaca dcntipes, caudal aspect. Fig. 476. Hydrotaea dentipes, mesal aspect. Fig. 477. Sarcophaga haeniorrhoidalis, caudal aspect. Fig. 478. Sarcophaga haemorrhoidalis. mesal aspect. Fig. 479. Stonwxys calcitrans, distal end. lateral aspect. Fig. 480. Stonwxys calcitrans, distal end> mesal aspect. Fig. 481. Lispa nasoni, distal end, mesal aspect. Fig. 482. Bombylius major, cross-section tbru pseudotrachea. (After Dimmock.) Fig. 483. Ochthera mantis, cross-section thru pseudotrachea. Fig. 484. Musca (Calliphora) voinitoria, cross-section thru pseudotrachea (After Dimmock.) Fig. 485. Musca (Calliphora) romitoria, an enlarged pseudotrachea. (After Dimmock.) Fig. 486. Oncodcs costattts, entire mouth-parts, caudal aspect. Fig. 487. Oncodes costatus, entire mouth-parts, lateral aspect. Fig. 488. Olfersia ardeae, distal end, lateral aspect. Fig. 489. Siinulium venustuni, cephalic aspect of the labrum. Fig. 490. Gastrophilus equi, entire mouth-parts, caudal aspect. Fig. 491. Gastrophilus equi, sagittal section thru mouth-parts. Fig. 492. Gastrophilus equi, entire mouth-parts, cephalic aspect. ILLIXOIS BlOl.OGICAI. MOXOGR.IPHS VOLUME 3 OncojM Simiiliiim Castrophilus Gasirophilus Gaslrophflw 487 489 490 491 492 I'KTF.RSOX Ill-.AD AXD MOUTH PARTS OF DIPTERA PI. \ Tl" XX 273] HEAD OF DIPTERA— PETERSON 103 PLATE XXI 104 ILLINOIS BIOLOGICAL MONOGRAPHS [274 \ EXPLANATION OF PLATE EpfPHARYNX AND HyPOPHARYNX AND ASSOCIATED PARTS Hypothetical type, lateral aspect. Tabanus giganteus, female, lateral aspect. Tabanus giganteus, male, lateral aspect. Tabanus giganteus, female, caudal aspect. Simulium venustum, female, lateral aspect. Siiiiuliuti! venustum, female, caudal aspect. Trichoccra biniacida, lateral aspect. Trichoccra biiiiacula, caudal aspect. Dixa clavata, lateral aspect. Dixa clavata, caudal aspect. Tipula bicornis, lateral aspect. Psorophora ciliata, female, lateral aspect. Psorophora ciliata, female, caudal aspect. Geranomyia canadensis, lateral aspect. Livinobia inuitatura, lateral aspect. Rhyphus punctatus, lateral aspect. Rhyphus punctatus. caudal aspect. Fig 493 Fig 494 Fig 495 Fig 496 Fig 497 Fig 498 Fig 499 Fig SCO Fig SOI Fig 502 Fig 503 Fig 504 Fig 505 Fig S06 Fig 507 Fig 508 Fig 509 ILLISOIS BIOLOGICAL MOSOGRAPHS VOLUME PETi;kSOX IIF.AI) AND MOUTH TARTS oi' HIPTI-KA \'\.\\V. XXI 275] HEAD OF DIPTERA — PETERSON 105 PLATE XXII 106 ILLINOIS BIOLOGICAL MONOGRAPHS [276 EXPLANATION OF PLATE Epipharynx and Hyfopharynx and Associated Parts Fig. 510. Rhabdophaga strohiloides, caudal aspect. Fig. 511. Rhabdophaga strobiloides, lateral aspect. Fig. 512. Sciara varians, caudal aspect. Fig. 513. Sciara varians, lateral aspect. Fig. 514. Pcriplaneta oricntalis, clypeiis, labrum, and epipharynx spread out, enta! aspect. Fig. 515. Melanoplus diffcrentialis, clypeus, labrum. and epipharynx spread out, ental aspect. Fig. 516. Gryllus pennsylvanicus, right-half of clypeus, labrum, and epipharynx, cephalic and caudal aspects. Fig. 517. Proiiiachus vcrtebratus, lateral aspect. Fig. 518. Promachus vcrtebratus, epipharynx and labrum, caudal aspect. Fig. 519. Promachus vcrtebratus, caudal aspect. Fig. 520. Leptis vertcbrata, lateral aspect. Fig. 521. Culicoides sanguisugus, lateral aspect. Fig. 522. Bibio fciiwratus, caudal aspect. Fig- 523. Bibio femoratus, lateral aspect. Fig. 524. Dolichopus bifractus, caudal aspect. Fig. 525. Leptis vertcbrata, caudal aspect. Fig. 526. Bibioccphala clcgantula. caudal aspect. Fig. 527. Bibioccphala clcgantula, lateral aspect. Fig. 528. Dolichopus bifractus, lateral aspect. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 Dolichop 524 PETERSON' HEAD A\D MOlTll PARTS (W DIPTERA PLATE XXIT 277] HEAD OF DIPTERA— PETERSON 107 PLATE XXIII 108 ILLINOIS BIOLOGICAL MOXOGRAPHS [278 EXPLANATION OF PLATE Epipharynx and Hypopharynx and Associated Parts Fig. 529. Psychoda albitcitiiis, lateral aspect. Fig. 530. Psychoda albipcunis, caudal aspect. Fig. 531. Chironoimis fcrrugincoziltatiis, lateral aspect. Fig. 532. Cliiroiwmus ferrugincorittatus, caudal aspect. Fig. 533. Psilocephala haemorrhoidalis, lateral aspect. Fig. 534. Psilocephala haemorrhoidalis, caudal aspect. Fig. 535. Mydas clavaius, lateral aspect. Fig. 536. Mydas clavatus, caudal aspect. Fig. 537- Scenopinus fcncstralis, caudal aspect. Fig. 538. Scenopinus fencstralis, lateral aspect. Fig. 539. Lonchoptera lutea, lateral aspect. Fig. 540. Aphiochaeta agarici, caudal aspect. Fig. S41. Lonchoptera lutea, caudal aspect. Fig. 542, Platypesa vchitina, caudal aspect. Fig. 542a. Platypc:a vclutina. lateral aspect. Fig. S43. Eidonchus tristis, lateral aspect. Fig. S44- Aphiochaeta agarici, lateral aspect. Fig. 545. Stratiomyia apicula, lateral aspect. Fig. 546. Stratiomyia apicula, caudal aspect. Fig. 547. Em pis clausa, lateral aspect. Fig. 548. Empis clausa, caudal aspect. Fig. 549. Exoprosopa fasciata, lateral aspect. Fig. 550. Exoprosopa fasciata. caudal aspect. n.LIXOlS BIOLOGICAL MOXOGRAPHS J-OLUME s 547 548 549 550 I'M 1 KRSUX ili;,\l) AXIJ .MUL' III PARTS OF DIPTKRA PLATF. XXIII 279] HEAD OF DIPTERA— PETERSON 109 PLATE XXIV no ILLINOIS BIOLOGICAL MONOGRAPHS [280 EXPLANATION OF PLATE EpiPHARYNX AND HyPOPHARYNX AND ASSOCIATED ParTS Fig. 551 Fig. 552. Fig. 553 Fig. 554. Fig. 555 Fig. 556. Fig. 557. Fig. SS8. Fig. 559. Fig. 560. Fig. 561 Fig. 562 Fig. 563 Fig. 564 Fig. 56s Fig. 566. Fig. 567 Fig. 568. Fig. 569, Fig. S70. Fig. 571 Fig. 572 Fig. 573 Fig. 574 Fig. 575 Fig. 576. Fig. 5: Fig. 578. Fig. 579- Calobata univitta, caudal aspect. Calobata univitta, lateral aspect. Sapromyca vulgaris, lateral aspect. Saf>roiiiyca vulgaris, caudal aspect. Chloropisca glabra, caudal aspect. Chloropisca glabra, lateral aspect. Chrysomysa deiiiandata, caudal aspect. Chrysoinyca deinandata. lateral aspect. Coelopa vanduzeii, caudal aspect. Coelopa vanduzeii, lateral aspect. Pipunculus cingulatus, caudal aspect. Pipunculus cingulatus, lateral aspect. Drosophila ampclophila, caudal aspect. Drosophila aiupelophila, lateral aspect. Borborus equinus, lateral aspect. Borborus equinus, caudal aspect. Borborus equinus, liypopharynx united with labium, caudal aspect. Chyroiiiya concolor, caudal aspect. Chyroiiiya concolor, lateral aspect. Lo.voccra pectoraiis, caudal aspect. Lo.roccra pectoraiis. lateral aspect. Euarcsta acqualis, caudal aspect. Euaresta acqualis, lateral aspect. Ochthera mantis, lateral aspect. Ochthera mantis, caudal aspect of the labrum. Ochthera mantis, caudal aspect of the epipharynx. Ochthera mantis, caudal aspect. Dcsmonictopa latipes, lateral aspect. Desnio-.uetopa latipes, caudal aspect. ILLINOIS BIOLOGICAL MONOGRAPHS VOLUME 3 Euvcsta EtuKsta Ochihcra Ochthet. Ochthcft 672 673 574 676 577 PETERSON HEAD AND MOUTH PARTS UI' DirXEKA PLATE XXIV \ 2811 HEAD OF DIPTERA— PETERSON 111 PLATE XXV 112 ILLINOIS BIOLOGICAL MONOGRAPHS [282 \ EXPLANATION OF PLATE Epipharynx and Hypopharynx and Associated Parts Fig. 580. Occothea fenestralis, lateral aspect. Fig. 581. Oecothea fenestralis, caudal aspect. Fig. 582. Sepsis violacea, lateral aspect. F'g- 583- Sepsis violacea, caudal aspect. Fig. 584. Tetanocera plumosa, lateral aspect. Fig. 585. Sphyracephala hrevicornis, lateral aspect. Fig. 586. Tetanocera plumosa, caudal aspect. Fig. 587. Eristalis tenax, caudal aspect. Fig. 588. Eristalis tenax, lateral aspect. Fig. 589. Heteroneura fiaviseta, lateral aspect. Fig. SQO. Heteroneura fiaviseta, caudal aspect. Fig. 591. Conops hrachyrhynchus, caudal aspect. Fig. 592. Conops hrachyrhynchus, lateral aspect. Fig. 593. Scatophaga furcata, lateral aspect. Fig. 594. Scatophaga furcata, caudal aspect. Fig. 595. Thelaira lexicosona, lateral aspect. Fig. 596. Thelaira leucozona, caudal aspect. Fig. 597. Hydrotaea dentipes, lateral aspect. Fig. 598. Hydrotaea dentipes, caudal aspect. Fig. 599. Stomoxys calcitrans, lateral aspect. l/Fig. 600. Musca doinestica, lateral aspect. Fig. 601. Musca domestica, caudal aspect. Fig. 602. Sarcophaga haemorrhoidalis, lateral aspect. Fig. 603. Sarcophaga haemorrhoidalis, caudal aspect. Fig. 604. Archytas analis, lateral aspect. Fig. 605. Archytas analis, caudal aspect. Fig, 606. Olfcrsia ardeae, lateral aspect. ILLIXOIS BIOLOGICAL MOSOGRAriiS VOLUME 3 604 606 PETERSOX HEAD AXl) MOUTH PARTS OF DIPTERA PLATE XXV ILLINOIS BIOLOGICAL MONOGRAPHS Vol.111 January, 1917 No. 3 Editorial Committee Stephen Alfred Forbes William Trelease Henry Baldwin Ward Published under the Auspices of the Graduate School Bt THE University of Illinois Copyright, 1917 By the University of Illinois Distributed May 5, 1917 STUDIES ON NORTH AMERICAN POLYSTOMIDAE, ASPIDOGASTRIDAE, AND PARAMPHISTOMIDAE WITH ELEVEN PLATES HORACE WESLEY STUNKARD Contributions (lom the Zoological Laboratory o( the University of Illinois ander the direction of Uenrj B. Ward. No. »4 THESIS Submitted in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy in Zoology in the Graduate School of the University of Illinois 1916 TABLE OF CONTENTS PAGE Introduction 7 Polystomidae _ 9 Historical Review of the Fainily g The Genus Polystoma i6 Anatomy and Histology of the Polystomidae 19 Polystoma orbiculare Stunkard 1916 31 Polystoma opacum Stunkard 1916 34 Polystoma megacotyle Stunkard 1916 37 Polystoma niicrocotyle Stunkard 1916 39 Polystoma coronatum Leidy 1888 40 Polystoma hassalli Goto 1899 42 Polystoma ohlongum Wright 1879 44 Key to the Species of the Genus Polystoma 46 Aspidogastridae 47 Aspidogaster conchicola von Baer 1827 48 Cotylaspis insignis Leidy 1856 49 Cotylaspis cokeri Barker and Parsons 1914 SO Classification of the Family 57 Parauiphistomidae 60 Historical Review of the Family „ 60 The Genus Alassostoma _ 64 Alassostonia magnum Stunkard 1916 66 Alassostoma parvum Stunkard 1916 69 Tlie Genus Zygocotyle _ _ 71 Zygocotyle ceratosa Stunkard igi6. 72 Classification of the Family 75 Relation of the Families to the Order 80 List of Xew Species 84 Bibliography 85 Explanation of Plates __ 91 2871 SORTII AMERICAS FOLVSTOMIDAE—STUXKARD INTRODUCTION The knoAvledgi' of the trematodes of North America is very scanty. Information at hand consists largely of brief and scattered papers and (;omprehensivc studies on the morphology of the larger groups are want- ing. Such studies are needed as contributions to the knowledge of adult forms, and it is apparent also that knowledge of the anatomy and tax- onomy of the adult is demanded in the solution of life history problems. This paper contains the results of a study on the structure and classification of North American representatives of the families Polystcmidae, Aspidogastridae, and Paramphistomidae. Because of cer- tain structural and developmental features these three families are of particular interest and importance not only in the taxonomy but also in the phylogeny of the trematodes. The Polystcmidae differ fi-om all other known Ileterocotylea in tliat tliey are endoparasitic ; the .\spidogastri- dae are both ectoparasitic and endoparasitic, develop both directly and by means of an intermediate host, and in the adult condition are pai'a- sites of both vertebrates and molluscs ; while the Paramphistomidae are the onl}- forms retaining a primitive postei'icr sucker. These facts are significant and it is probable that further study into the structure and life history of these forms will throw considerable light on the general jiroblems of development and taxonomy of the trematodes. During the past three years the writer has made parasitological examinations of over three hundred North American fresh-water turtles. These compri.se sixteen species collected from widely scattered localities. For assistance in securing tliis material, grateful acknowledgments are due Dr. N. A. Cobb of Washington, D. C. Professor A. W. Orcutt of Denison University, Professor \V. E. Burge of the University of Illinois. Professor J. E. Aekert of Kansas State Agricultural College, and Professor \V. W. Cort of Macalcster College. Tlie material of Alassostoma parviim was collected and turned over to me by Mr. T. B. Magath. A type specimen of Pohjstoma coronaium Leidy from the U. S. National Museum was placed at my disposal for study. The work was begun at the suggestion of Professor Henry B. Ward and carried on under his direction. Part of the material used in the investigation came from his private collection, and for this material as well as for criticisms and suggestions in the course of the work the writer wishes to express his appreciation. 8. ILLINOIS BIOLOGICAL MONOGRAPHS [288 All the forms described in tliis paper were studied as toto mounts; where sufficient material was available sections were made, and many were studied alive. The importance of the study of the living specimens can not be overemphasized as the best method of tracing the excretory system. Also, by observing the living animal as it moves, it is possible to measure the extent of normal variation in form that occurs in a single specimen as different shapes are assumed concomitant with the move- m.ents of the animal ; in forms with such soft bodies and variable shapes, a study of preserved material alone has in many eases given false con- ceptions concerning morphological relationships of organs and systems. In toto mounts a support under the eoverglass is necessary to prevent it from flattening and distorting the normal shape of the aspidogastrids and to avoid crushing the caudal disc of the poly.stomes. For the stain- ing of specimens to be mounted in toto, better results were obtained by the use of carmine than by hematoxylin stains. For staining sections the method that proved most valuable was to use the hematoxylin stains for differentiating the nuclear elements and various plasma stains for counterstaining. XORTH AMERICAX POLVSTOMlDAESTrXKARD POLYSTOMIDAE HISTORICAL REVIEW OF THE FAMILY In 1758 Roesel von Rosenhof described and figured a "leeeir' from tlie urinary bladder of the frog. Tliis is regarded as identical with the well known European parasite of the urinary bladder of the frog, described by Frohlich (1791) as Linguatuln integer rimum. M. Braun (1792) described Planai-ia uncinulata from the urinary bladder of the green water-frog and his description is so specific tliat there can be no doubt that he had the same form described by Frohlich the previous year. Zeder (1800) founded the genus Polystoma to contain the three species, Linguatula intcgerrimum Prohlicli which he rechristened Pohj- stoma ranac, P. serratum, and P. pinguicola. According to Stiles and Hassall (1908) the type was clearly intended to be P. ranae = Plaiiaria uuciiivlata, and altho Braun had described the form correctly with the suckers and hooks at the posterior end of the body, Zeder erroneously stated in liis characterization of the geuus that the suckers were at the anterior end. P. serratum had been designated by Frohlich (1789) as type of the genus Linguatula and P. pinguicola had been designated by Treutler (1793) as type of the genus Ile.xathyridium. That Zeder was in error in including these species in the genus Polystoma was demon- strated by later studies. However Rudolphi (1809) retained them in the genus Polystoma and listed three other species: P. taenoides Rud., P. drnii<:iilatiim, Rud., and P. vcnarum (Treutler 1793) Zeder 1803. Among these species, it is probable that Treutler's description was of an artifact rather than a parasite, and tlie other two have been i-cmoved to the Linguatulidae. Pohjstoma thijnii was described from the gills of Scomber fhynnus by Delaroche (1811). Rudolphi (1819) renamed this species P. dupUca- tum and added a new species P. occllatum from tlie tliroat of Emgs europa. This species is regarded as identical witli that described by Kuhl and Hassalt (1822) from the nasal cavity of Ilalirhrli/s atra. P. logiginis was described by delle Chiaje (1823) from LoUgo vulgar i,<. Blainville (1828) oriented the polystomes correctly and transferred P. iniegerrimum, P. orellatum, and P. thgnii to a new genus Ile.xacotyle, naming H. thynii as type. According to the rules of zoological nomen- clature, however, the genus Polystoma must be retained. Kuhn (1829) 10 ILLIXOIS BIOLOGICAL MOXOGRAPHS [290 described P. appcndiciilatain from Squalus catuhis. Dujardio (1845) transferred Didibothrium crassicaudatum Leuck. 1835 = Diplobothrium itrmatum Leuck. 1842 to the genus Polystoma, and listed as additional species, P. duplicatum, P. pinguicola, P. ocdlatum, P. intcgcrrimum, and P. appcndiculatum. Diesing (1850) named P. loliginis and P. appcndic- vlatum as types of new genera Solenocotyle and Onchocotj-le. He re- moved P. armatum to the genus Didibothrium Leuck. and retained in the genus Polystoma ouh' the species P. infigcrrimum and P. occlkitum. Tlie genus Polystoma together with the genera Tetrastomum, Gry- ])orhyneluxs, Hexathyridium, Notocotyle, Aspidocotyle, and Aspidogas- ter were included by the same author in the tribe Polycotylea. In his revision Diesing (1859) reduced the trematodes to the rank of a tribe and divided the group into three subtribes: Acotylea, Coty- lophora, and Plectanophora. The second of these subtribes he subdi\'i- ded into three families : Monocotylea, Tricotylea, and Polycotylea. The last of these corresponds almost identically with his former tribe Poly- cotylea. He rejected Gryporhynchus, and added the genera Ancyroceph- alus, Plagiopeltis, Heptastomum, Onchocotjde, Cyclocotyle, and Solen- ocotyle. In the family Polycotylea he recognized two subfamilies: Aplacocotylea with the suckers set directly in the body, and Placocotylea with the suckers set in a median posterior plate. In the latter he in- eluded the genera Onchocotyle, Polystoma, Cyclocotyle, Aspidocotyle, Aspidcgaster, and Solenocotyle. Then followed the great work of van Beneden (1858) with an ex- perimental demonstration of the "direct" development of the many- suckered ectoparasitic trematodes, and the "indirect" development of the distomes. For the.se two groups he proposed the names Monogenea and Digenea. In the former he recognized two families: the Tristomi- dae with a single posterior sucker, and the Polystoraidae with several posterior suckers. In the Polystomidae he included the genera Polystoma, Diplozoon, Octobothrium, Axine, Onchocotyle, Calceostoma, and Gyro- dactylus. Later van Beneden and Hesse (1863) made the genera Octocotyle ( = Octobothrium), Udouella, and Gyrodactylus types of new families, thus increasing the number of families to five. Many additional genera, both old and recently described, were now for the first time placed with the Monogenea. But in the family Polystomidae these authors retained only two genera, Polystoma and Erpocotyle; and in the genus Polystoma was listed onlj' a single species, P. intcgerrinium. Taschenberg (1879) reverted to the earlier classification of van Beneden and adopted the division of the moncgeuetic trematodes into two 291] XORTH AMERICAX rOLYSTOMIDAE—STUSKARD 11 groups Tristomeae and Polystomeae, which he regarded as families Under the Polystomeae as subfamilies he listed Polystomidae, Octoboth- ridae ( = Oetocotylidae), Gyrodaetylidae, and the new subfamily Micro- eotylidae ; the latter including Microcotyle, Axiue, Gastrocotyle and the entirely unlike genera, Aspidogaster, Cotylaspis and Aspidoeotyle. To the Polystomidae he added the genera Onchocotyle and Diplobothriura, and in the genus Polystoma included the two species P. intcgcrriinum and P. occllaium. In regard to the previously mentioned forms St. Remy (1891) fol- lowed the famih' and subfamily divisions of Taschenberg, tho adding new genera to each of the subfamilies and removing Aspidogaster, Coty- laspis, and Aspidoeotyle from the Microcotylidae. To the Polystomidae, Wright and Maeallum had added the genus Sphyrauura, and in the genus Polystoma were listed the new species P. oblongum Wriglit and P. coro- nafum Leidj'. Increased knowledge of the trematodes disclosed so many exceptions to their classification according to life history that Monticelli (1892) proposed a new arrangement of the group, based on morphological char- acters. To contain the forms previously classed as Monogenea, he pro- posed the suborder Heterocotylea. He raised the Monocotylidae and Gyrodaetylidae from subfamily to family rank, making five families in the Heterocotylea. In the family Polystomidae he retained the sub- families Polystominae, Oetoeotyliuae, and Microeotylinae of former authors. So far as the Polystomidae are concerned, the synopsis of Pratt (1900) does not differ from that of St. Remy and Monticelli. Later Monticelli (1903) worked out a new classification of the Heterocotylea, separating the forms on the basis of differences in the ad- hesive apparatus. He an-anged the families in two tribes, Oligocotylea and Polj'cotylea, the former containing the forms with few suckers and the latter those with many suckers. This division he says is not of great systematic importance but may be of practical value in the identiiicatiou of families. In the Oligocotylea he included the families Tristomidae, Slonocotylidae, Udonellidae, Calceostomidae, Gyrodaetylidae, and Dico- tylidae ; and in the Polycotylea the families Polystomidae, Oetocotylidae, Hexacotylidae, Platycotylidae, Pleurocotylidae, and Microcotylidae. Among these the Udonellidae, Oetocotylidae, and Microcotylidae are raised from subfamily to family rank, and the Calceostomidae, Dicoty- lidae, Hexacotylidae, Platycotylidae, and Pleurocotylidae are new fam- ilies. The family Polystomidae contained the single genus Polystoma with the species P. integerrhnum, P. occllutuin, P. oblongum, P. coro- natum, and P. hassalli. 12 ILUXOIS BIOLOGICAL MOXOGRAPHS [292 Diseiassing the classification of Monticelli, Odlmer (1912) stated that he considered the number of suckers of secondary importance and the system based on them therefore lacking in fundamental systematic sig- nificance. Accordingly he rejected the work of Monticelli, and using the older classification of Monogenea, divided the forms within the group on the basis of differences in the female reproductive diicts. He discussed the relationship of the ducts of the female genital system in various trema- tode and cestode genera, and stated that he was convinced as was claimed by Stieda that Laurer's canal of the trematodes should be regarded as homologous with tlie vagina of the cestodes. Intervening authors, Looss (1893), Goto (1894) and several other writers, had considered Laurer's canal of the Malacocotylea as homologous with the genito-intestinal canal of certain Ileterocotylea, and not with the vagina of the cestodes. Odh- ner argued that Laurer's canal was the primitive vagina of the trema- todes and that there had been a change of vaginal function from this canal to the teimiinal part of the uterus, with the resulting degeneration of the former duct. It now served in his opinion only to carry ofi' excess spermatozoa, together with yolk and shell substance not used in the for- mation of the eggs. Pie adds that there is no evidence on which to base an explanation of the transfer of the seat of vaginal function from Laurer's canal to the terminal part of tlie uterus ; it must only be accepted as a fact. According to Odhner, in the group of monogenetic trematodes, two very different structures are included under the term vagina. One pres- ent in the Tristomidae, Monocotylidae, and Gyrodaetylidae opens to the exterior on the left side of the ventral surface, and at the inner end is enlarged to form the seminal receptacle. This tube he considered homo- logous to the vagina of the cestodes and Laurer's canal of the digenetic trematodes. The other structures which he did not consider homologous to this true vagina were the ducts of the Octocotylidae, Polystomidae, and Microcotylidae, which function as vaginae and open into the vitelline collecting ducts. These are paired and open to the surface either ven- trally, laterally, or dorsally. Contending that they had arisen siti generis, he proposed for them the name "ductus vaginalis." Considering tlie question of whether the paired or unj^aired condition of these ducts was primitive, he argued that originally the duct was unpaired and opened ventrally; that the opening became divided and the duct split, therefore the Y-shaped duct of Kajonchocotyle must be considered as a stage in the development of the paired condition of the ducts. A further separation would give the lateral openings of Polystoma. In the Microcotylidae the openings have migrated dorsally and fused producing a single dorsal tube. Odhner could find no homo- logue for the genito-intestinal canal and since he maintained that it was 293J XORTIl AMEKICAX POLYSTOMIDAE—STUXKARD 13 not homologous witli Laurer's canal, coiu-huled that it had arisen sui generis. On the basis of these differences in the female genital duets he divided the Mouogenca into two suborders : Monopisthocotylea and Poly- opisthoeotylea. The former is characterized by the absence of tlie gonito- intpstinal canal, the presence of a "true vagina" and a single pos- terior organ of attachment; the latter bj- the presence of the genito- intestinal canal, "ductus vaginalis," many posterior adhesive organs, and the absence of a "true vagina." In the Monopisthocotylea he included the families Tristomidae, jMonocotylidae, Udonellidae and Gyrodactylidac ; and in the Polyopisthocotylea the families Polystomidae, ]\Iicrocotylidae and Octoeotylidae. He pointed out that by tlie removal of the genus Sphyranura, the Oligocotjdea, the first of Jlontieelli's tribes agrees entirely with his suborder llonopisthocotylea. In the second of Mouticelli's tribes, however, the Diclidophorinae, to- gether with the genera Dactylocotyle and Hexacotyle, should be re- moved from the Octoeotylidae and placed with the Microcotylidae, since the}- more nearly agree with the latter forms in internal structure. The next year Odhner (1913) reaffirmed his idea of the homologj' of the vagina of the cestodes and Laurer's canal of the distomes, but explained therewith that his denial of the homology of the genito-intestinal canal and Laurer's canal had been based on an error of Cerfontaine in describing an unpaired vagina as present in the genus Dac- tylocotyle. On examination of this genus ho had found that a "true vagina" was absent, and concluded that tlie "true vagina" of the Mono- pisthocotylea which lie had homologized with Laurer's canal was never j)resent together with the genito-intestinal canal. From this he decided tliat tlie "true vagina" was homologous with the genito-intestinal canal ami therefoi-e with Laurer's canal. Now maintaining the homology of the "true vagina'' and the genito-intestinal canal he is i;i my opinion obliged to dismiss the presence or absence of the genito-intestinal canal as a basis of difference between his suborders, and explain why in one group this canal opens to the exterior on the ventral side of the body and in the other opens into the intestine. His homology of the "true vagina" and the genito-intestinal canal is a most serious error since it would invalidate tlic distinguishing feature which separates the two suborders. I propose to sliow that the organ which functions as a vagina is liomologous in all tlie monogenctic trcmatodes, and that there can be no division of the group on the basis of difl^'erences suggested by Odliner. In fact, the work of Odhner is based on an incorrect assumption and false homologies. Starting with the assumption that Laurer's canal is homologous to the vagina of the cestodes, he has missed the truth in his 14 ILLIXOIS BIOLOGICAL MOXOGR.IPHS [294 entire discussion and when at a loss to explain a structure has derived it sui generis. His later paper (1913) admitting the homology of Laurer's and the genito-intestinal canals corrected one mistaken contention, but his separation of the female copulatory ducts into a true vagina and "eanalis vaginalis" seems entirely \vithoi;t foundation. There is no evi- dence to support the idea that the single vagina is not homologous to the paired vaginae. In fact, Odhner described the paired vaginae as arising by the division of a single unpaired tube, probably ventral in position. He derived this tube sui generis, and cited no reason why it is not homo- logoiis with the ventral unpaired vagina of the Monopisthocotylea. Fur- ther he gives no means of distinguishing between the two. Looss (1893) presented a strong argument to prove that Laurer's canal is not a vagina, nor homologous to the vagina of the cestodes. Goto (1894) reveiewed the literature up to that date and gave a careful and detailed study of the eanalis gc nito-intcstinalis. Making a very clear and comprehensive analysis of the question and summarizing evidence from a wide study of ectoparasitic forms, he concluded that the genito-intestinal canal and Laurer's canal are homologous and that neither are homologous with the vagina of the Monogenea. He showed that in the group there is a perfect series of vaginae from a truly paired to a truly unpaired condition. He discvissed the idea of Braun who re- garded the presence of a single vagina as the result of a simple atrophy of one of the originally paired vagina, with the conclusion that the rela- tions of the ducts "point strongly to the view that the impaired vagina has been formed by the union and subsequent displacement of the ori- ginally paired vaginae, and not as Braun supposes by the atrophy of one of them." In the present study, the histological character and the relative posi- tion and relationships of the ducts of the female system support the con- tention of Looss and Goto that Laurer's canal is homologous with the genito-intestinal canal, and affords no evidence that these ducts have any further homologue. A review of the literature and the study of the ducts in the three families discussed in this paper has convinced me that Laurer's canal is homologous to tlie genito-intestinal canal ;and the vagina of the Monopisthocotylea is homologous with the originally single, sub- sequently paired, and secondarily fused vaginae of the Polyopisthocotylea. It makes no difference whether the single or paired condition is regarded as primitive. Given a single unpaired vagina as described by Odhner for the Monopisthocotylea ; by a division of the external part and subse- quent lateral migration of the openings, the paired vaginae of the Poly- opisthocotylea are explained. These duets entering the body from the sides, lying parallel with the vitelline ducts and discharging into the 295] XORTH AMERIC.LX POLYSTOMIDAE—STiWKARD 15 same cavity, became fused at their internal ends with the vitelline ducts and this union continued outward to the location where the vitelline ducts turn toward the follicles and the vaginae branch off to open to the exterior. The advantage of a single duet over two ducts lying side by side is obvious, and the fusion of two parallel ducts is not uncommon in other groups. With a further dorsal migration of the opening of the vaginae there would be a separation of the vitelline and vaginal canals and a dorsal fusion of the vaginae would give the single dorsal vagina of Octobothrium, Axine, and Microcotyle. The earlier fusion of the vitel- line and vaginal canals would retard the secondary fusion of the internal ends of the dorsal vaginae and this explains the single dorsal pore and internally paired vagina of Axine hctcroccrca which is used by Odhner as an argument supporting his idea that in the jMonogenea two different structures are included under the term vagina. I agree with Odhner that the seminal receptacles of Sphyrauura are homologous to the paired vaginae of Polystoma, and that this furnishes a splendid example of the change whereby the terminal part of the uterus has assumed the copulatory function. It may be that further specializa- tion in this direction, due to the endoparasitic habit and self fertilization, may explain the absence of the vagina of the distomes. It now remains only to account for the absence of the genito-in- testinal canal in the Monopisthocotylea. Odhner stated that this struc- ture is homologous M'ith Laurer's canal, and in his (1912) paper called attention to the fact that Laurer's canal is a "rudimentary organ" which serves no essential function. The vestigeal character of Laui-er"s canal is believed in by most writers — Looss, ]\Ionticelli, Brandes, Goto, etc. This structure is entirely lacking in some distome groups and in others is represented by a blind sac opening from the ootype. Since the genito-intestinal canal is admittedly homologous to Laurer's canal and the latter is kno^vn to be a vestigeal structure, it appears reasonable to suppose that it has degenerated in the ^Monopisthocotylea. There is a possibility that the Monopisthocotylea instead of having lost a genito-intestinal canal may have arisen fi'om a group of the Tur- bellaria which had no homologous structure, but this explanation seems very improbable. Haswcll (1907) described in certain Australian poly- clads a tube which formerly had been considered an accessory or dorsal vagina but which in certain forms opened into the intestine. The pres- ence of this genito-intestinal canal in polyclads, he says, "strengthens the contention, so ably supported by Goto, that the genito-intestinal canal and not the vagina of the Heterocotylea is the equivalent of the Laurer's canal of the Malacocotylea." The absence of the genito-intestinal canal in the ^lonopisthocotylea 16 ILLINOIS BIOLOGICAL MONOGRAPHS [296 is undoubtedly a feature of distinct taxonomic importance, and the work of Odhner is an advance step in the formation of a natural system and the final classification of the monogenetic forms. Since the arrangement of ilouticelli, based on the character of the adhesive apparatus, so nearly agrees with that of Odhner which in reality is based on the presence or absence of a genito-intestinal canal, it appears that both these features are of large importance in the taxonomy of the group. Present evidence is insiifificient to decide which is of greater significance. Further study may show that there is complete agreement in classifications based on both features. Odhner (1912) argued that the removal by Monticelli of Sphyranura from the family Polystomidae on the basis of the difference in number of suckers was not justified. As previously stated, the writer agrees with Odhner that the seminal receptacles of Sphyranura are homologous M'ith the vaginae of Polystoma, and the agreement in type of genital ducts in- dicates a closer relationship between these genera than is assigned in the system of Monticelli. Sphyranura undoubtedly should be placed with the Polyopisthocotylea. There are, however, wide and fundamental differ- ences between it and the genus Polystoma, and while future researches may discover intermediate forms which will make it possible to include them with certainty in a single family, for the present such a grouping is hardly justified and the two families should be retained, altho the name Dieotylidae of Monticelli does not conform to the rules of zoological nomenclature. THE GENUS POLYSTOMA The family Polystomidae as considered in this paper contains only the genus Polystoma. The members of this genus are widely distributed, species having been described from all the continents except South America. The species are not only widely distributed geographically, but also vary widely in type of host and in locatisn within the host. They are parasitic in the urinary bladder of frogs and toads and on the gills of frog larvae, and also infest the urinary bladder and phar3-ngeal cavity of many species of turtles. The structure and development of Polystoma integerrinium has been investigated by Stieda (1870), Zeller (1872 and 1876), Willemoes-Suhm (1872), Halkin (1902), Goldschmidt (1902), and Andre (1910). Zeller (1876) described two forms of P. integerrimum, one which became ma- ture in the urinary bladder of the frog, and the other which became ma- ture on the gills of the frog tadpole. These two forms of the parasite show wide differences in size and internal structure. The form which becomes mature in the urinary bladder is much larger, has a lobed testis, external vaginae, and a long coiled uterus which contains many 297] XORTH AMERICAN POLYSTOMIDAE—STUXKARD 17 eggs. The form maturing on the gills of the tadpole has a spherical testis, lacks external vaginae and a long coiled uterus, and has a small uterine cavity in which a single egg develops. Ilalkin and Goldschmidt have investigated the early stages in this form, but the writer has been unable to find any reference to work on the later larval stages. The findings of Zeller are so unusual that one is led strongly to suspect he confused two different species. The descriptions of P. oceUatum by Rudolphi (1819) and Kuhl and Hassalt (1822) are very brief; that by Willemoes-Suhm (1872) contains one plate, and Looss (1885) figured only the structures at the distal ends of the excretory tubules. The description of P. oUongum Wright (1879) contains sufficiently detailed information for a specific diagnosis and is illustrated by three figures. Stafford (1905) reported P. ohlongum from the palate of Chry- semys picta and the same location in Chehjdra serpentina, but since "Wright originally described the species from the urinary bladder of Aromochelys odoratus, Braun reviewing Stafford's article considered the form from the oral cavity as a different species. The form described by Leidy as P. oblongum, was reinvestigated by Goto (1899) and proved to be a different species from that described by "Wright, but the material he reports was in such a poor state of preservation that renewed study was impossible and so the form must remain unknown. Leidy's (1888) description of P. coronatum is so bi-ief that it is al- most valueless ; a type specimen mounted as a toto pi'eparation has been available for the present study and many additional points of structure are added to the original description. P. Jwssalli was described by Goto (1899) from the urinary bladder of Cinosternum pcnnsyhmnicum and has been collected by the writer from the iirinary bladder of Aromochelys odoratus, A. carinatus, and Chelydra serpentina, as well as from Cinosternum pennsylvanicum. Additional data correct and siipplement the description of Goto. Johnston (1912) described P. iulliense from the urinary bladder of two species of Hyla from New South Wales, Australia. Beauchamp (1913) described P. alluaudi from an unknown batrachian from the lower prairies of Kinangop, Africa; the material was collected by the African expedition of Alluaud and Jeannel. Stewart (1914) described P. kachu- gae from the urinary bladder of the water tortoise, Kachuga lineata, at Lucknow, India. In the genus Polystoma present evidence supports the validity of the following described species listed in the order of description : P. integerrimum Frolich 1791. From the urinary bladder of frogs and toads and the gills of frog larvae ; Europe. 18 ILLIXOIS BIOLOGICAL MONOGRAPHS [298 P. ocellatum Rudolphi 1819. From the throat and nasal cavity of Emys europa and Halichelys atra; Europe. P. ohlongum Wright 1879. From the urinary bladder of Aromo- chclys odoratus; North America. P. coronatum Leidy 1888. From the fauces of the terrapin ; North America. P. hassalli Goto 1899. From the urinary bladder of Cinosternum pennsylvanicum, Aromochclys odoratus, A. carinatus, and Chelydra ser- pentina; North America. P. bullicnsc Johnston 1912. From the urinary bladder of Hyla phyllochros and H. Lcsueurii; Australia. P. alluaudi, Beauchamp 1913. From an unknown batrachian ; Africa. P. kachugae Stewart 1914. From the urinary bladder of Kachuga lineata; India. In the present work evidence is submitted to justify the inclusion of the following new species: P. orhiculare Stunkard 1916. From the urinary bladder of Pseu- demys scripta and Chrysemys marginata; North America. P. opacum Stunkard 1916. From the pharynx of Trionyx fcrox and Malacoclemmys lesueurii; North America. P. mcgacotyle Stunkard 1916. From the mouth of Chrysemys mar- ginata; North America. P. microcotyle Stunkard 1916. From the mouth of Chrysemys mar- ginata; North America. With the exception of P. integerrimum, the members of the genus are very rarely found and the number of individuals discovered is very small. Wright described P. ohlongum from two specimens; Leidy, P. corona- tum. from four specimens; Johnston had sixteen specimens of P. iulliense; Beauchamp described P. alluaudi from a single specimen ; Stewart had only two specimens of P. kachugae. The writer had only a limited num- ber of individuals of any species ; P. microcotyle was described from a single specimen; P. orbicidare from nine specimens; P. opacum and P. megacotyle each from three specimens. Becavise of the limited amount of material, it has been impossible to attempt special technique to differ- entiate the various organ systems, and the descriptions are therefore in- complete in certain particulars. The general morphological features are however described in sufficient detail that clear specific diagnoses can be made, and in certain instances the finer structure and histology of the organs has been described. 299] XORTH AMERICAX POLYSTOMID.IE—STUXKARD 19 ANATOMY AND HISTOLOGY OF THE POLYSTOMIDAE The species that have been included in the genus Polystoma show a much wider range of structural variation than is usually present in a natural genus. There are wide differences in the character of digestive and reproductive systems, and variation exists in the type of adhesive apparatus. There is wide variation in size; P. intrgcrrimum, the largest known species measures up to 12 mm. in length, and P. hassalli is only 1.3 to 2 mm. in length. The width is one-third to one-fifth of the total length. All the worms that have been included in this genus have a flattened, elongate oval body which at the posterior end bears a large ventral muscular disc or cotylophore. The body is more or less pointed at the anterior end and at the posterior end may or may not have a con.striction just before the attachment of the caudal disc. As in all trematodes the shape is subject to considerable variation as the animal elongates and contracts. Locomotion is accomplished by attaching the anterior sucker and then bringing the caudal disc forward ; as a result of the terminal attachments and the "looping" method of progression, the dorsal line of the body is more or less arched and the ventral surface is concave. In certain species at the openings of the vaginae on the lateral or ventro- lateral margins of the body, there are prominent swellings, the ''Seiten- wiilste " " of Zeller. These structures are not present in any of the known North American species. Organs of Attachment. — The caudal disc bears on its ventral face the chief organs of attachment. These consist of suckers and hooks, the former arranged in pairs, three suckers on each side of the median line. The two posterior suckers are close together and those of the middle pair are separated by a considerable distance, while the anterior pair may or may not be near each other. In all previously reported forms except P. alliiaudi, the anterior suckers are separated by a considerable distance, giving the disc the shape described b}' Leidy as cordiform (Fig. 27). In the single specimen of P. alluaudi described by Beauchamp, both the caudual and cephalic suckers are separated, while those of each side are contiguous. In P. orbiculare the anterior suckers are in the same close proximity as the caudal pair, and each sucker of the disc is separated from the two adjacent to it by uniform distances, making a perfect circle of suckers (Fig. 1). In the six species described by the writer these suckers are complicated structures, set more or less deeply in the paren- chyma of the caudal disc. Their structure, character of insertion, mus- cular attachments, and relation to the surrounding tissue indicate that they are protrusible and retractile, and in fact such movements may be observed bv watching the live worm. 20 ILLIXOIS BIOLOGICAL MOXOGRAPIIS [300 The suckers are cup shaped (Fig. 34), and in all the species describ- ed in this paper are constructed on an elaborate cutieular framework. According to Zeller the sucker forms as a ridge around a larval hooklet and later sinks into the parenchyma, and this method of origin explains the cutieular covering of the external and internal surfaces of the cup. Riuming across between these cutieular membranes, tliere ai'e short re- fractive fibers which constitute the mass of tlie wall of the sucker (Fig. 35). Wright and Macallum (1887) describing similar fibers in the walls of the suckers of Sphyranura say, "Instead of the substance of the sucker being formed of muscular fibers disposed in three directions, and capable of modifying the shape of the cavity, as in the distomes, it is not possessed of contractility in Sphyranura (and probably in Polystoma), and is formed of prismatic Abel's, rather of a supportive than a muscidar character, arranged perpendicularly between the concave and convex limiting membranes of the suckers." Goto (1894) described similar fibers in the suckers of Axine, Microcotyle, Oetocotyle, Diclidophora, Hexacotyle, and Onchocotyle and considered them to be more of an elastic than a contractile nature. They are, he states, difi'erent from the ordi- nary muscular fibers of the body and from those of the suckers of the Tristomidae and Monocotylidae, as well as from those of the anterior sucker of Onchocotyle, both in optical characters and in reaction toward staining fluids. The structure of the suckers in these forms and their mode of operation are discussed by Goto at considerable length, but as the suckers he described are constructed on a different type of cutieular framework from that present in the genus Polystoma, obviously the type of suctorial action is different. In all the species described in this paper, the fibers which form the walls of the posterior suckers are similar to those described by Wright and Macallum and Goto; the cutieular framework is also flexible and elastic, but is of a different type from that described by Goto. In the polystomes investigated by the writer, with the exception of P. integer- rimum, the sucker consists of three sections or zones which may be desig- nated as basal, intermediate, and external or distal (Fig. 36). The ex- ternal part or rim of the sucker is supported by numerous cutieular rods formed by the thickening at regular intervals of the cutieular lining. These rods are bent outward, their curvature maintaining the flare of the rim of the sucker. Distally thej^ terminate just inside the rim of the cup and basally they are continuous with and are processes from a band of cuticula which passes around the sucker and separates the external and intermediate portions. In toto preparations this band appeal's to be divided into sections that are almost square, each with a circular area in the center that increases and decreases in size as the focus is changed. 3011 XORTH AMERICAX POLYSTOMIDAE—STUSKARD 21 Sections show that the cuticular lining of the sucker is folded outward against the convex wall with which it is fused, thus interrupting the con- tinuity of the fibrous wall (Fig. 35). The two sides of this invagiuated cuticular sac or ring are fused at regular intervals, leaving small pockets alternating with the places of fusion. These small openings in the cuti- cular band are conspicuous by reason of their different refractive index and show very plainly with a dark field illumination as the square or rectangular sections with the circular areas in the center (Fig. 3i). There is apparently no relation between the number of these sections in the cuticular band and the number of cuticular thickenings which serve as supports of the external section. The middle section of the sucker extends basally from the previously described cuticular band to a somewhat similar evagination of the cuti- cular lining into the wall of the sucker, but this evagination does not ex- tend to the external cuticular covering of the sucker and onlj- partially divides the fibrous wall. This middle or intermediate portion of the sucker is supported by thickenings of the cuticular lining, processes that extend peripherally from the cuticular band which passes around the sucker at its base. These supporting ridges are not arranged at regular intervals and they are much fewer in number than the cuticular rods which support the external section. They are often branched, tho not more than a single bifurcation was observed. The basal portion of the sucker is circular, similar in structure to the portions previously described ; it has internal and external limiting mem- branes with fibers extending between. At its center the cuticular and fibrous wall is interrupted and there is the structure described by John- son (1912) as the connective tissue plug, which appears as a central disc or button, and to which the retractor muscles are attached. This central disc has thickened cuticular edges and bears the larval booklet. Figure 44 illustrates the method of operation of the suckers. Muscles are at- tached to the external wall of the distal and intermediate portions of the sucker and the contraction of these muscles retracts the two external zones, with the accompanying protrusion of the basal part. Whether the small hooks at the bases of the suckers are functional is doubtful. As previously described, the cuticular supports do not extend quite to the external margin of the sucker, leaving a soft plastic edge which can be applied all the waj- around even on an irregular surface. With the con- traction of the muscles attached to the basal disc, a vacuum is produced and forms a powerful means of adhesion. Since the walls of the sucker are not contractile and the suckers vary only slightlj' in size in a single species, the size of the suckers has been used by the writer as a character for determining specific identity. 22 ILLIXOIS BIOLOGICAL MOXOGRAPHS [302 A cuticular framework similar to that present in Polystoma was described by Wright and Macallum for the suckers of Sphi/ranura osleri. They say: "As the wall of the sucker is itself destitute of contractility, another arrangement exists for modifying the shape of the cavity. Its walls is really divided into three concentric zones, which by special ex- trinsic muscles can be worked independently. The two circular lines which separate these zones, are marked by an infolding of the investing membrane, which forms a sort of joint, permitting an independent move- ment of the zones." The collection of Professor Ward contains a single series of sections of P. intcgcrrimvm from Germany, and in this specimen the type of skeletal structure just described is absent. Figure 45 shows the charac- ter of the suckers in this form. The caudal disc typically bears eighteen hooks. Sixteen are similar in size and shape, arranged as follows : six in a row between the anterior suckei's, one situated inside each sucker at the base, and four between the two posterior suckers. In addition to these hooks there is a pair of great hooks, several times the size of the small hooks, between the two posterior suckers. The shape of these hooks and their arrangement are shown in Figures 37 to 43. In many cases there is only one pair of the small hooks between the caudal suckers ; in such eases in addition to the great hooks there is a third pair, similar in shape to the great hooks and intermediate in size between the great and small hooks. The sixteen small hooks are present on the caiulal disc of the larva before the suckers are formed and are called larval booklets by Wille- moes-Suhm (1872), but Zeller (1876) says: "Die sechszehn kleinen Hakchen mit iliren Oesen, welche die Haftseheibe angehoren und welche bei der Polystomum larva so ausscrordentlich deutlich zu erkennen ist, sind nieht, wie Willemoes-Suhm meint, nur 'Larvalorgane'. Sie werdeu nicht abgeworfen, sondern sind wie ich auf das bestimmeste wiederholen muss, bei der erwachsenen Thiere noch sammtlieh vorhanden, sehr beweglich und gewiss nicht ohne Bedeutung fiir ein festeres Anheften." Jolinston (1912) in the description of P. huUicnsc says: "Four larval booklets are present in a row- on the ventral surface near the posterior edge of the disc or cotylophore. I have been able to find no trace either in the living worms or the fixed material, of the larval booklets which P. intcgerrimum and other species bear near the anterior edge of the disc. There is a small anchor shaped hook in the base of each sucker. All these hooks either disappear as the animal increases with age, or very readily become detached. In only one out of sixteen specimens have the whole four posterior booklets been present ; and in only two others were any booklets at all to be seen. In all other specimens no liooklets could be made out." 303] XORTH AM ERIC AX POLYSTOMIDAE—STUXKARD 23 In my own material I find that the larval hooklets are invariably present in the bases of the suckers, but of the other larval hooklets, us- ually several are absent and often those present are so arranged that it is difficult to see how they could function in attachment. Those at the an- terior edge of the caudal disc are seldom regularly arranged, and in many eases (Figs. 37 to 43) are in such irregular and unusual positions with reference to each other that the use of one would interfere with the action of the others. The great hooks are invariably present in the species in which the caudal disc is cordiform in shape, i. e., where the two anterior suckers are separated by a distance exceeding that between the two posterior suckers. In the species P. alliunidi and P. orbiculare the disc is circular and the great hooks are absent. Usually the cordiform disc is wider and the cir- cular disc is narrower tlian the body. At first it seemed possible to separate the genus into two subgenera, one in which the disc is circular and the great hooks are absent and another with a cordiform disc and great hooks present, but there seems to be no such clear line of separation. In P. orbiculare a large number of chitinous spicules are present on the disc, some between the suckers and the others in the central area of the disc. In P. opacum the disc is intermediate in shape ; it is difficult to determine whether it is circular or cordiform, and the great hooks are present altho they are not more than half the size of those in other species (Fig. 40). In P. hassalli the disc at times may be circular and the great hooks are strongly developed (Fig. 31). Bodij Covering. — The body is covered with a non-cellular, unarmed cuticula, which is turned in at the external openings of the various sys- tems. It does not have a uniform appearance but is traversed by lines which extend perpendicular to the surface of the body. Musculature. — The musculature consists of the dernio-muscular sac, the muscles of the adhesive apparatus, and dorso-veutral strands with much-branched fibers wliieh traverse the body at irregular intervals. The muscles of the body wall consist of an external circular layer, an interme- diate layer of diagonal fibers, and inside the latter, bundles of longitud- inal fibers. In all the species studied, the inner longitvidinal fibers are more strongly developed than either of the other layers. Stieda (1870) in P. intcgcrrimum did not distinguish between the two external muscle layers and described only two layers of muscles, an outer layer of an- nular fibers, some of which were not exactly circular and crossed each otlier, and an inner layer of longitudinal fibers. Zeller (1876) was in error when he described the diagonal fibers as inside the longitudinal layer in P. integerrimum. The arrangement of the muscles of the body wall in Polystoma is the usual condition in the Heteroeotylea, and a 24 ILLIXOIS BIOLOGICAL MOXOGRAPHS [304 similar arrangement has been described in Calicotyle, Axine, Nitzscliia. Tristomum, Octobothrium, Temnocephala, Microeotyle. Octocotyle, and Monocotyle. In Diclidophora Goto (1894) described an additional layer of longitudinal fibers between the circular and diagonal layers. He states that in Onchocotyle and Hesaeotyle the circular fibers seem to be en- tirely lacking. In the genus Polystoma there are strong sets of longitudinal fibers near the median line on the ventral side of the body. They could be traced antei'iad only to the testis. Posteriad they pass into the caudal disc and together with fibers from the body wall are inserted on the sides and in the bases of the bothria. Muscle strands from both sides of the body pass to each of the suckers (Fig. 29) and smaller groups of fibers from each sucker to each of the others. In addition to the dorso-ventral muscles which extend between various points of the body wall, there are other fibers from the body wall to the internal organs. Mcscnchyma. — The mesenchymal tissue of the body does not show a differentiation into ectoparenehyma and endoparenchyma as described by Brandes (1892) and other writers; it is not of a uniform character, but presents differences in appearance at different points in the same specimen. It may take the form of compact cellular tissue, or of vacuo- lated cells, or there may be large vacuoles apparently between cells, or the cellular sti-ucture may be entirely lacking, there being only a reticu- lum of fibrous tissue. The parenchyma is traversed by many muscle strands, and the dorsal and lateral regions are occupied by the enormous- ly developed vitellaria (Figs. 19, 23). Alimentary System.— The digestive apparatus consists of a terminal anterior or oral sucker, a pharynx, a short esophagus and a bifurcate in- testine. The oral sucker (Fig. 6) is not fully homologous with that of the distomes. There is no external limiting membrane, branched muscle fibers passing from the inside lining of the sucker to the body wall. Posteriorly it is limited and separated from the body parenchyma by special strands of fibei-s which pass from the body wall to the Mall of the digestive tube and are attached there just anterior to the pharynx. A contraction of these fibers causes the constriction between the anterior sucker and the body parenchyma which is sometimes seen. Longitudinal muscle fibers from the body parenchyma penetrate this posterior boun- dry of the anterior sucker and pass to the wall of the sucker. Annular muscles, situated just inside the cutic\ilar lining, pass around the sucker from side to side. Situated among the muscle fibers there are large secretory cells. Johnston described the structure as a weakly developed or incipient oraj sucker. The anterior sucker, pharynx, and esophagus are lined with cuticula continuous with that of the external surface of the body. 305] NORTH AMERICAN POLYSTOMIDAE—STUNKARD 25 The pharynx is approximately spherical, altho various states of con- traction influence its shape to some extent. It does not lie directly in the long axis of the body but obliquely, tlie lumen extending from the some- what ventral anterior opening from the oral sucker to a more dorsal pos- terior opening into the esophagus or intestine. In certain species it is composed of two portions, (Figs. 6, 33) tho both are enclosed in the same external capsule. In the anterior portion there are many strong annular fibers and this part probably acts as a spliineter. altho there are also radial fibers which extend from the external limiting membrane to the cuticula of the lumen. In the posterior part the annular fibers are con- fined almost entirely to the external region and a small central zone (Fig. 25). The muscle fibers are branched and non-nucleated. Scattered among the fibers in the posterior part there are large nuclei, each with a deeply staining nucleolus and surrounded by a granular or flaky area that is continued by a fine duct traceable by the presence of the same granular substance and leading to the lumen of the pharynx. Goto described somewhat similar nuclei in the pharynx of Diclidophora and regards them as remnants of the cells that have produced the muscle fibers. The writer is inclined to the view that in Polystoraa the granular substance is a secretion. No extra-esophageal glands were observed, but that the secretion of the pharyngeal cells is salivary was not demon- strated. A short esophagus may be present in certain species (Fig. 6) but in most cases the pharynx appears to open directly into the intestine at the juncture of the right and left ceca. There may be a short median or paired lateral pockets of the intestine extending anteriad from the junc- tion of the ceca. There is wide variation in the type of the intestinal diverticula. In P. iniegcrrimum the ceca are much branched and the.se branches ramify thru the body and the caudal disc (Fig. 45). In P. alluaudi the ceca oc- cupy the same location but are merely lobed and have no secondary branches, tho they are united posteriorly. In P. buUiense, according to Johnston, "a diverticulum from the buccal cavity i-iuis backwards, ven- tral to the pharynx, and for a distance equal to its length forming a me- dian unpaired buccal pocket." In all other known species there is a simple bifurcate intestine, the ceca terminating just anterior to the caudal disc. In two specimens of P. hassalli, however, the ceca are connected posteriorly; in one of them tlie ends of the ceca are continuous and in the other there is a connection some distance anterior to the ends of tlie ceca (Fig. 30). The walls of the diverticula are composed of a delicate fibro-membranous tissue upon M-hich rests the digestive epithelium. Tlie einthelial layer consists of columnar cells whose nuclei lie near the fibro- 26 JLLIXOIS BIOLOGICAL MOXOGRAPHS [306 membranous sheet and which have large, rounded, often vacuolated bodies extending: irregularly into the canal. The protoplasm of the cells is granular. Excretory System. — In this family as in all Heterocotylea, there are two excretory pores, situated on tlie dorsal surface about midway between the median line of the body and the lateral edge of the worm, near the level of the caudal margin of the pharynx (Fig. 27, 33). These open from vesicular expansions, which when filled are almost spherical and when empty have folded walls. The descending collecting duct originates in the region of the pharynx from the fusion of smaller ducts and passes posteriad to the region of the caudal disc where it turns eephalad and continues as the ascending collecting duct to open into the excretory vesicle. Both the descending and ascending ducts receive smaller branches at irregular intervals; at the caudal end of the body a canal joins the tubes of the two sides and a similar connection exists between the descending ducts just anterior to the pharynx. From this anterior communicating canal a branch enters the anterior sucker near the median line. The excretory vesicles are lined with a thin layer of cuticula con- tinuous with that of the external surface of the body and the collecting ducts and accessory branches have a fibro-membranous wall in which nuclei are occasionally embedded. In P. intcgcrrimum, Zeller described many connections of the collecting ducts of the two sides thru anastomoes of their smaller branches. He also described cilia on the walls of the col- lecting ducts. Looss (1885) described the excretory system of P. ocella- turn. He says the collecting ducts are not ciliated throuout, but only in occasional areas, and describes cilia in the capillaries. These capillaries are long and at the distal end are verj' much coiled. In this coiled part the capillary is divided so that two flame cells discharge into each coil and are emptied by a single capillary. The caliber of the excretory ves- sels is very minute, and altho varying somewhat as a result of distention, lacunar expansions were not observed. Because of the limited amount of material, much of which was received in a preserved condition, no at- tempt was made to trace the excretory system in living worms of this family. The vitellaria completely obscure the excretory ducts in toto preparations. The secondary ducts are so small and so often collapsed that it is impossible to follow their continuity with certainty in sections. Nervous System. — The morphology of the nervous system of P. integerrimum was described in detail by Andre (1910). He described a supra-esophageal brain from -which three pairs of nerves pass anteriad and three pairs posteriad. In another paper (1910a) he gave a detailed description of the eyes of P. integerrimum. In the present work no special study of the nervous system was made and no new facts were adduced. 307] XORTH AMERICAN POLYSTOMIDAE—STUNKARD 27 Male Reproductive System. — The testis is a much branched structure in P. kachugac; in P. integerrimuni it is lobed, and in the other known species it is oval or spherical. It is situated near or slightly anterior to the middle of the body. A duct designated an internal vas deferens was described in P. integerrimuni by Zeller, but Ijima (1884) traced tlie true relations of this tube and showed that it passes from the ootype to the in- testine. Goto (1894) proposed the name canalis gcnito-intcstinalis for this structure which is discussed in a later section. The vas deferens arises from the dorso-cephalic margin of the testis and passes dorsad and anteriad. It extends dorsal to the ootype, between the dorsal margins of the ovary and uterus to the level of the genital pore where it turns ven- trad and enlarges to form the seminal vesicle (Fig. 13). From the semi- nal vesicle a duct pas.ses thru the cirrus sac, opening into the genital atrium (Fig. 26). The vas deferens is small and has a tibro-membran- ous wall, and the seminal vesicle has a lining of columnar epithelium. The cirrus sac is composed of an external mviscular wall enclosing a mass of parenchymous tissue which surrounds the ejaculatoiy duct. This sac is very small in P. intcgerrimiim and P. hassalli. Ventrally it opens into a common genital atrium (Fig. 26). The ejaculatory duct terminates in the genital papilla, which when retracted is surrounded by a deep depres- sion. In the musculature between this depression and the wall of the cirrus sac are embedded the roots of the genital hooks. When the hooks are retracted there is a shallow depression between them and the wall of the sac. With the contraction of the wall of the cirrus sac the genital papilla and the circle of genital hooks are extruded thru the pore. In most of the species the hooks are sickle shaped with the points project- ing outward, and with muscles attached to the outside of the hook at the juncture of the root and shank. These muscles undoubtedly serve as a fulcrum, and the extrusion of the papilla rolls the hooks outward bury- ing their points in the cuticula lining the wall of the vagina of the copu- lating worm (Fig. 24). In P. alluaudi Beauchamp described three genital hooks, P. integerrimum has eight, and other species sixteen, thirty-two, and forty. In P. hassalli the genital hooks are small, straight and have a wing like process at the middle. Zeller described a prostate gland in P. integerrimum, consisting of masses of large cells situated around the cirrus, and traced ducts from these cells to the lumen of the ejaculatory duct. Johnston in P. bulliense says, "Two laterally placed, small groups of gland cells represent the prostate." The statement of Zeller that a gland is present aroiuid the cirrus of P. integerrimum is certainly correct. In the species described in this paper, a similar gland is present in the parenchyma around the genital sinus. The cells (Fig. 12) are globular or pyriform, stain deeply 28 ILLIXOIS BIOLOGICAL MONOGRAPHS [308 and possess a distinct nucleus and nucleolus. Their ducts could not be traced to the ejaculatory duct but in many cases appear to lead to the body wall near the margin of the genital sinus. In P. orbiculare, P. opacum and P. megacotyle the cirrus sac is large and many nuclei are present around the ejaculatory duct in the dorsal part of the sac. These nuclei are large, with distinct nucleoli, and are surrounded by a deeply staining area of granular or flaky substance, but no cell boundries could be made out. Female Reproductive System. — In all known species but one, the ovary is oval or comma shaped. In P. kachugae it is described by Stew- art as a "curved sausage-shaped organ, the curve forming all but a com- plete circle. The fundus is somewhat bulbous." This structure is usually not more than one half the size of the testis, is situated a short distance anterior to that organ, and in a given species may lie on either side of the body. In all the species studied by the writer it is comma shaped, the larger part is ventral, anterior, and lateral, and terminal region is dorsal, posterior, and mesal. The ova are formed in the large part and the ovary is divided into zones of growth, ova of increasing size being present in each succeeding zone (Fig. 23). In the species described in this paper the vitellaria consist of masses of follicles occupying the dorsal and lateral regions of the body. Each follicle consists of several cells which may vary much in appearance ; the difference is due to the phase of secretory activity of the cells. In the peripheral part of the gland the cells are usually small, with granular or flaky protoplasm, a distinct nucleus and nucleolus; whereas those more centrally located may be two or three times their size, the extra-nuclear area being either vacuolated or filled with droplets of a yellow substance (Figs. 19, 20). In some cells the secretory droplets are scattered uni- formly thruout the cell. The presence of the material in the cells often renders the body so opaque that the diverticula can not be seen. The glandular secretion is apparently identical with that which forms tlie shell of the egg, and this observation further confirms the statement of Gold- schmidt (1909) that the so-called vitellaria secrete the shell of the egg. Small ducts from the follicles (Fig. 11) unite and discharge into longi- tudinal collecting ducts. These extend along the sides of the body, later- al to the ceca and dorsal to the excretory tubules ; on either side of the body there is an anterior and a posterior branch which unite just behind the level of the ovary and the common duct discharges into the external end of the vitello- vaginal canal. In P. hiilliense, Johnston reports: "The lateral vaginal swellings are formed by a large number of pajDillae, per- forated by fine canals, which after a very short course, open into a fairly wide sperm reservoir, situated one on either side, just under the swell- 309] XORTH AMERICAS POLYSTOMIDAESTUNKARD 29 iiigs. From these reservoirs, a wide vaginal tube on either side runs backwards and inwards, to open into the anterior lateral yolk duct." A similar condition is described and figured by Zeller for P. intcgerri- mum. In all other species in which the structure has been described, the vaginae are open funnels leading mediad and dorsad from their openings on the ventro-lateral surface of the body, and uniting just below the in- testine with the common vitelline ducts to form the viteUo-vaginal canals. The cuticular lining of the vaginae is very thick and in the parenchyma around the vaginae there are large cells of secretory type (Fig. 24). The ^^tello-vaginal canals lead medially and unite, either forming a duct which discharges into the ootj-pe (Fig. 32) or opening separately into the ootype (Figs. 3, 16, 24). From the ovary the o\aduct passes posteriad and ventrad, opening into the ootype. Immediately anterior and dorsal to the opening of the o^-iduet, there branches from the ootj'pe a small tube which after a some- what twisted double loop opens into the intestine of the side in which the ovarj- is situated. This genito-intestinal canal has been the source of much controversy and its presence or absence is the diagnostic feature of Odhner's two groups of monogenetic trematodes. Mehlis' gland, the shell gland of earlier authors, is never largely developed and is difficult to find in some specimens where it is represented by a few nuclei in the parenchyma around the ootype. Zeller for P. integerrimum and John- ston for P. huUiense described prominent "shell glands", and Stewart for P. kachugae described "a group of glandular cells found at the same transverse level as the ovary, but on the opposite side of the midline. They appear to be connected with the corresponding vagina, but their function is obscure." Since they are in the precise location of the Mehlis' gland, one is led to suspect that Stewart was confused in regard to tlie connections and relations of this group of cells, altho in in- dividuals of other species studied by the writer, there are groups of large glandular cells in the parenchyma surrounding each vagina. The ootype is continued by a tube which passes anteriad on the op- posite side from the ovary, and which leads to the uterus. Previous writers have called this tube the oviduct and Johnston (1912) says, "From the ootype, the oviduct runs forward to a point in front of the ovary, whence it bends sharply backwards and runs in a straight course close to the ventral surface, almost to the level of the cotjdophore, where it opens into the wide uterus. ' ' The use of the term o^'iduct for the tube leading from the ootype to the uterus is confusing and objectionable. Looss (1899) says, "Der Theil des weiblichen Leitungswegen, der den Keimstock mit deui ootj-p verbindet, ist der oviduct oder Keimgang," and this terminologj' is found in general use thruout the literature. In a large number of trematode genera the ootype opens directly into the 30 ILLIXOIS BIOLOGICAL MONOGRAPHS [310 Uterus. In the Polystomidae however, there is a definite specialized tube leading from the ootype to the uterus. This duet is not homologous with the oviduct, it is separated from that duct by the ootype, and further, in the specimens examined by the writer the histological character of the two are not precisely the same. The epithelial lining of the oviduct is of the flattened type, and that of the second duct more columnar. Such a duct is present in many cestode genera and is called the uterine duct. The same name is proposed for the tube leading from the ootype to the uterus in the Polystomidae, altho with the understanding that its use is independent of the question of homologies of the female ducts in treraa- todes and cestodes. In P. hulliensc the uterine duct opens into the uterus not at the end but on the side, and there is a posterior uterine pocket. The uterus ex- tends as a wide elongated sac from the extreme posterior end of the body to the common genital sinus. In P. alluaudi the intracecal area is occu- pied by the uterus and eggs are present almost as far posteriad as the caudal union of the ceca. In P. integerrimum there is a long icterus which extends in many loops anterior to the ootype, and contains a large number of eggs. In all other kno\^^l forms, the uterus is situated at the level of the ovary on the opposite side of the body, and contains a single large egg or embryo. Zeller (1876) described a similar condition for the ectoparasitic form of P. integerrimvm. Figure 14 shows a very early embyro of P. orbicidarc and Figure 23 a much later stage of development in P. megacotyle. No shell is present in the former case, altho it may have been lost in sectioning. There must be some provision for the growth of the embryo and the shell can not be rigid during the uterine period. Where the oviduct arises from the ovary, at its union with the ootype, and at either end of the uterine expansion sphincter muscles pro- duce short contracted portions of the tube. In all the species studied by the writer, with the exception of the vitelline tubules, all ducts of the female system have a fibro-membranous wall and an epithelial lining, which in the ootype, uterine duct, and uterus consists of tall columnar cells with distinct boundries and single nuclei. Describing tlie ei^ithelial cells lining the ootype in certain other mouogenetie forms Goto (1894) says that because of their appearance and reaction to stains he strongly suspects theii' glandular nature, but since a shell gland is present he can not understand their function. In certain species of Polystoma Mehlis' gland is much reduced or absent, and in these forms the cells of the epithelial lining of the ootype appear to be secretive (Figs. 8, 9). This agrees with the present conception that the vitellaria secrete the shell substance and Mehlis' gland the fluid in which the eggs are suspended. The genital pore is situated on the ventral surface in the median 311] XORTH AMERICAN POLYSTOMIDAE—STUXKARD 31 line, just posterior to the bifurcation of the digestive tract. It opens from a common genital sinus (Figs. 13, 26) into which the uterus dis- charges and thru which the cirrus is extruded. The opening from the uterus into the genital sinus is posterior and ventral, while the cirrus sac opens into the dorsal part of the atrium. When the two specimens of P. opacum from Trionyx ferox were placed in a watch glass, they soon came in contact and immediately started copulation, the cirrus of each worm was inserted in the right vagina of the other, and the two worms attached to each other, both with the anterior suckers and those of the caudal disc that coidd be brought in position for adhesion. Attempts to separate the worms failed, so an effort was made to fix them in the copulating condition, but they separated on the application of the killing flvud. This explains the statement of Johnston : ' ' On one side only, in the specimens sectioned, was the vaginal tube filled with sperms; that on the other side was empty." Benham (1901) and Mac Galium (1913) state that copulation in poly- stomes has been observed only by Zeller. POLYSTOaiA ORBICULAKE Stunkard 1916 [Figures 1 to 14] The material of this species consists of six specimens from the urinary bladder of Pseudemys scripta from Raleigh, North Carolina, one specimen from the urinary bladder of Chrysemys marginata from Chicago, Illinois, and two specimens from the urinary bladder of Chry- semys marginata from Creston, Iowa. The body is an elongate oval, slightly more pointed anteriorly than posteriorly, and in two of the specimens with slight indentations of the body walls at the vaginae and at the posterior margin of the anterior sucker. These worms (Fig. 1) varied in length from 2.7 to 3.75 mm. and in width from 0.9 to 1.2 mm. The caudal disc is circular, 0.8 to 1.07 mm. in width, and bears the six suckers arranged symmetrically in a circle. The suckers are approximately 0.3 mm. in diameter, and are separated by regular equal intervals. No hooks could be found on the caudal disc with the exception of the single minute larval booklet in the base of each sucker. These are 0.016 mm. in length and could be seen only under favorable conditions. The anterior sucker (Fig. 6) is 0.25 to 0.27 mm. in length and 0.37 to 0.42 mm. in width. It opens into the pharynx, a spherical struc- ture 0.24 to 0.28 mm. in diameter. There is a short esophagus visible in sagittal sections altho it is not distinguishable in toto preparations. The ceca meet anteriorly in a wide curve and extend as simple tubes 32 ILLIXOIS BIOLOGICAL MONOGRAPHS [312 almost to the posterior end of the body. They have no branches and terminate blindly. In caliber they vary from 0.04 to 0.116 mm. The testis is spherical or oval, visually slightly longer than broad, and measures 0.29 to 0.39 mm. in width and 0.36 to 0.5 mm. in length. It is near or slightly anterior to the middle of the body. The sperm duet arises at its anterior margin and, lying dorsal to the ootype, passes anteriad. In front of the ovary it turns ventrad and expands into the seminal vesicle. At the outer end of the seminal vesicle the duct is encircled by a sphincter muscle, and then known as the ejaculatory duct passes thru the cirrus sac to open into the genital atrium (Figs. 3, 13). The cirrus sac is almost spherical, and consists of an external muscular capsule filled with parenchjonatous tissue enclosing a central canal. In the dorsal part of the sac there are radial muscles passing from the wall to the central duct, and among these fibers a few large nuclei. More ventrally there are sets of muscles developed around the central duct and these are connected to the wall of the sac. Externally the central canal terminates at the apex of a papilla which is separated by a deep depression from the muscivlar ring that bears the hooks of the genital coronet. This conical muscular ring is protrusible and is separated from the wall of the cirrus sac by a second depression. The invaginations on either side of the genital coronet allow for the extru- sion of the coronet of hooks with the genital papilla on the contraction of the wall of the cirrus sac, while the muscles attached to the central canal and the muscular ring bearing the genital hooks serve as retract- ors. The genital coronet consists of sixteen hooks, similar in size and shape ; they have an external sickle-shaped part or shank which turns outward and a root or basal part of about the same length embedded in the musculature (Figs. 2, 13). The basal part is straight and hollow and the internal end is bifurcate. It bears many fine cuticular processes which are particularly prominent near its union with the shank. In the body parenchyma around the terminal part of the cirrus sac there are large iiuicellular glands (Figs. 12, 13). The ovary is lateral and may be situated on either side of the body. It is 0.1 to 0.25 mm. anterior to the testis. It is ovoid in shape, with the larger part in which the ova are being formed anterior and ventral, and the oviduct arising from the dorsal posterior region. In sections it appears to be marked into zones, with larger and fewer cells present in each succeeding zone. It is 0.1 to 0.148 mm. in width, 0.14 to 0.185 mm. in length and in one specimen cut in cross sections 0.175 mm. in depth. The oviduct arises as a very small tube and immediately expands (Fig. 3). This expanded portion extends posteriad and ventrad and by means of a short constricted tube opens into the ootype, a specialized region 313] XORTH AMERICAN POLVSTOMIDAE—STLWKARD 33 of the female duet where the vitello-vagiual canals are received and the geuito-intestinal canal is given off. The genito-intestinal canal twists in a double loop and then opens into the intestine of the side upon which the ovary is located (Fig. 10). The vaginae are ventro-lateral in position and open to the exterior by funnel shaped mouths. The vitel- laria occupy the dorsal and lateral regions of the body; they extend anteriad to the pharynx and posteriad to the caudal disc. There is a strand of follicles across the dorsal side of the body just behind the pharynx, and then the follicles are entirely extracecal in the field ante- rior to the testis; posterior to the testis the vitellaria overlie the eeca and extend to the center altho they are scanty along the median line. Ventrally the vitellaria are entirely extracecal. Collecting ducts run longitudinally, laterad of the ceca ; and just below the cecum of either side the common vitelline ducts formed by the union of the anterior and posterior longitudinal ducts unite with the internal ends of the vaginae to form the vitello-vaginal canals. These canals open directly into the ootype, one on either side, and are thus continuous, forming a canal thru the body from side to side. Mehlis' gland is represented by i.iany nuclei which lie in the parenchyma around the ootype and uterine duct. This latter duct passes anteriad and laterad on the opposite side from the ovary ; it is smaller than the ootype in diameter and the epithe- lial lining is lower. After a slight expansion it is constricted and then opens into the uterus. The uterus contained a single egg or embryo. Figure 14 shows a morula-like mass of cells found in one specimen ; in the other specimens there were large spherical eggs, each enclosed in a yellow shell. They vary from 0.21 to 0.24 mm. in diameter. The excretory system shows no departure from the typical form and while it can not be completely followed in sections, the larger ducts occupy the characteristic positions. The descending collecting duets arise in the region of the anterior sucker and pass posteriad, lying lat- eral and ventral to the ceca. They wind back and forth in short curves and at the posterior end of the body turn anteriad and pass in the same winding course to the excretory vesicles. Both descending and ascend- ing ducts receive small branches at irregidar intervals. The excretory pores are lateral and dorsal, at the level of the bifurcation of the intestine (Fig. 7). This species agrees with P. alluaudi in shape of caudal disc and absence of great hooks, but differs from that species in type of uterus, number of hooks in tlie genital coronet, and in the character of the intestinal diverticula and testis. P. oriiculare agrees with P. hassalli in the number of genital hooks, but the hooks are different in size and shape ; P. hassaUi lias the great hooks of the caudal disc well developed 34 ILLIXOIS BIOLOGICAL MONOGRAPHS [314 whereas they are absent in this species. In certain particulars P. or- hicidarc resembles P. opacum, but the two species have different num- bers of hooks in the genital coronets ; they differ also in the relative size of caudal suckers. The great hooks of the caudal disc are present in P. opacum. The two species differ also in that one is parasitic in the urinary bladder and the other in the oral cavity. POLYSTOMA OPACUM Stunkard 1916 [Figures 15 to 21] Two worms of this species were obtained from the esophagus of a single specimen of Trionyx ferox from Newton, Texas, and another from the esophagus of Malacoclcmmys lesueurii from the same region. These trematodes were the same color as the lining of the esophagus and so firmly attached that they were removed only with great difficulty. The worms (Fig. 15) measured 4, 3.75, and 3.25 mm. in length and 1, 0.85 and 0.8 mm. respectively in width. The body has an elongate oval outline, is flattened dorso-ventrall.y, and observed in living condi- tion, shows great variations in shape. In an extended condition it nar- rows at either or both ends, and the contracted form may be not more than half the length when extended, and broadly oval or quadrate in outline. The caudal disc is slightly wider than the body in the mounted specimens, measuring 1.09 and 1.21 mm. in width while each sucker is approximately 0.4 mm. in diameter. The suckers have a chitinous skele- tal framework, as is described in the generic discussion. In the external meridinal band there are thirty-two divisions, which number corre- sponds to the number of hooks in the genital coronet. The suckers are arranged in a circle, altho the anterior pair are separated by a distance slightly exceeding that between the posterior pair. Between the anterior suckers there are many chitinous spicules, and in one specimen two of the larval booklets. Chitinous spicules are present on the sides of all the suckers and over the ventral surface of the disc. Between the posterior suckers there are three pairs of hooks, viz. two pairs of the small larval hooks and one larger pair, but the great hooks are relatively much smaller than the corresponding hooks in other species in which they are present (Fig. 40). The larval booklets are 7 to 9/* in length and the great hooks are 75;u, in length. The chitinous spicules present on the disc have no definite arrangement and their points may stand in any direction ; the three larval hooks between the anterior suckers of one specimen have no definite relative position and their hooks point in different directions; those at the posterior edge of the disc are set 315] NORTH AMERICAN POLYSTOMIDAESTVNKARD 3S in a row at more or less regular intervals and their hooks all point backward. The cuticular covering of the body is about 14ju in thickness, and on the contraction of the body is thrown into minute folds and furrows. The anterior sucker is oval, 0.2 to 0.22 mm. in length and 0.23 mm. in width. It opens into the pharynx (Fig. 18), a spherical structure . 0.3 mm. in width. There is a broad nerve commissure crossing the an- terior part of the pharynx M'hich contains large ganglion cells. Prom this dorsal commissure a nerve passes venti'ad on either side of the pharynx. Tlie digestive tract is of the simple triclad type, the pharynx is fol- lowed by a short esophagus, 0.17 mm. in length in the sectioned worm, and the diverticula extend as simple tubes almost to the posterior end of the body. They are about 0.15 mm. in diameter and terminate blindly, dorsal to the middle pair of suckers (Fig. 21). The eeca are lateral but close together, separated by only 0.2 to 0.25 mm. They have the usual fibro-membranous coat and epithelial lining, and were empty in the sectioned individual. The testis is spherical or slightly longer than broad in well extended specimens. It is slightly anterior to the middle of the body and is com- posed of a large number of lobes or strands of cells, compacted and enclosed in a membranous capsule. Cells with the chromatin of their nuclei in all stages of division and mature spermatozoa were observed in sections. The sperm duct arises at the anterior dorsal margin of the testis and curves dorsad and eephalad. Anterior to the uterus it turns ventrad and expands to form the seminal vesicle. From the seminal vesicle a small ejaculatory duet leads through the cirrus sac and opens into the common genital sinus. The ovary is ovoid or comma shaped, situated a short distance anterior to the testis, and in all three specimens is located on the left side of the body; but since in other species it may lie on either side, it is probable that the examination of a larger number of individuals would show specimens with the ovary on the right side. In dorsal view it is from 0.16 to 0.2 mm. in length and 0.08 to 0.12 mm. in width, while in the specimen that was sectioned it is 0.08 mm. in widtli and 0.3 mm. in depth. The oviduct arises at the dorsal posterior margin and curves posteriad, mediad, and ventrad where it opens into the ootype. The vitello-vaginal canals open separately into the ootype, just ventral to the origin of the genito-intestinal canal. The latter duct passes latcrad, then dorsad and anteriad, turns mediad almost to the median line of the body, then dorsad and laterad, and opens into the intestine of the side in which the ovary is located. The uterine duct passes to the right 36 ILLISOIS BIOLOGICAL MOXOCRAPHS [316 sight of the bodj-, then dorsad and anteriad where it opens into the uterus. Mehlis' gland is present altho not well developed, and the cells are scattered along the uterine duct as well as around the ootype, altho they are not so numerous in the former as in the latter location. The vaginae open to the surface on either side at the ventro-lateral margins of the body, at the level of the posterior margin of the ovary (Fig. 16). On either side the inner ends of the vaginae iinite just below the ceca with the common ducts from the vitellaria to form the vitello-vaginal canals. These open separately and directly into the ootype. The vitel- laria consist of large compact follicles, underlying the entire dorsal surface of the body from the pharynx to the caudal disc, except the region over the ovary. The vitellaria are reduced and only a few folli- cles are present in the region over the testis and they are entirely absent in a circular area over the ovary. Ventrally the vitellaria do not extend mediad of the ceca. The vitellaria are so extensively developed that they obscure the internal structures and render the body opaque, and this character suggested the name of the species. Common collecting ducts run longitudinally along the body lateral to the intestinal diverticula and these discharge into the vitello-vaginal canals as previously de- scribed. In each of the specimens there is a single large egg in the uterus, and in the one sectioned the uterus extends eephalad of the genital pore and to a point 0.03 mm. from the bifurcation of the intes- tine. The eggs are broadly oval, 0.25 mm. long by 0.2 mm. wide. The shell is yellow, refractive to light, and apparentl.y composed of the same substance that occurs in small droplets in the vitellaria. The uterus and cirrus sac open into the genital sinus; the opening of the cirrus is anterior and dorsal to that of the uterus. The common genital pore is situated in the median line, about 0.12 mm. caudad of the bifurcation of the intestine. Embedded in the wall of the cirrus sac and with their points forming the so-called coronet, the genital hooks in appearance suggest the corolla of a flower. There are thirty-three of these hooks in one mounted specimen and thirty-two in the other. In entire length they measure 0.05 mm., the shank or projecting part com- prising about half the total length. P. opacum agrees with P. alluaudi and P. orhkulare in shape of caudal disc, but P. alluaudi has but three spines in the genital coronet, and a long post-ovarian uterus which contains many eggs. P. orhkulare has a larger anterior sucker, smaller caudal suckers, a smaller pharynx, fewer vitelline follicles, and only half as many hooks in the genital coronet. P. opacum differs from P. coronatum and P. microcoiyle in the shape of the caudal disc and in the reduced condition of the great hooks of the disc. 317] XORTH AMERICAX POLYSTOMIDAE—STUXKARD 37 POLYSTOMA MEGACOTYLE Stunkaid 1916 [Figures 22 to 26] The material of tliis species consists of three specimens from the mouth of Chrysemys nmrginata from Crestou, Iowa. One worm was cut into cross sections and the other two mounted as stained toto preparations. These worms (Fig. 22) have an elongate ovoid shape. Widest in the region just anterior to the caudal disc, they gradually become nar- rower anteriorly, and posteriorly they taper rapidly to a caudal tip which is set in the antero-eentral part of the caudal disc. The worms are 2.5 to 2.7 mm. long and 0.71 to 0.78 mm. in width. The caudal disc is eordiform and the suckers are so large that they slightly overlap each other. The suckers are arranged in about four-fifths of a circle around the lateral and caudal margins of the disc. Measurements thru the disc from side to side at the level of the cephalic suckers are from 1 to 1.4 mm., thru the middle pair 1.2 to 1.8 mm., and thru the caudal suckers 0.68 to 0.7 mm. The disc bears the characteristic armature of liooks. Across the anterior margin there are three larval booklets in one speci- men and four in the other, but their arrangement is not regular or definite and their position would indicate that tliey do not function in attachment. In the specimen reproduced in Figure 22 the two hooks of the right side have their points almost together and their bases apart. In the bases of the suckers there are small larval booklets, and one pair similar in size and shape between the two caudal suckers. Also between the posterior suckers (Fig. 41) there is the pair of great hooks and a pair of hooks the same shape as the great hooks and intermediate in size between the great and larval hooks. The hooks measure in length : lar- val 0.017 mm., great hooks 0.116 mm., and the pair intermediate in size 0.058 ram. The cuticular covering of the body is approximately 5/^ in thick- ness on the dorsal and 3 to ^jj. in thickness on the ventral surface. It is turned in at the external openings and lines the digestive tract to the bifurcation. The anterior sucker is set off from the remainder of the body by a slight constriction. It is oval, its longest axis crosswise of the body, somewhat flattened posteriorly, and measures 0.28 mm. in length by 0.35 to 0.42 nun. in width. It is followed by the pharynx (Fig. 25) which is 0.35 to 0.38 mm. long, 0.38 to 0.44 mm. broad, and in the .sec- tioned worm 0.34 tliick. No esophagus was observed; the ceca meet anteriorly in a wide curve and extend almost to the posterior end of the body. They are 0.06 to 0.11 mm. in diameter, and have an epithelial 38 ILLINOIS BIOLOGICAL MOXOGRAPHS [318 lining 0.017 to 0.035 mm. in thickness set upon a fibro-membranous base. The vitellaria are so thick that the diverticula can not be traced in toto preparations. The testis is situated near the center of the body; it is spherical or oval, 0.28 to 0.33 mm. long, 0.33 to 0.38 mm. wide, and in the sec- tioned worm 0.28 mm. thick. The course of the vas deferens and the character of the male organs are similar to those in the previously described species. The genital coronet contains thirty-six hooks in one and fortj-^-two in the other toto preparation. They are similar in size and shape, have a straight basal portion with bifid end which is embedded in the wall of the cirrus sac, and a sickle shaped shank which projects into the genital atrium. The basal portion is the same length as the shank and each part measures 0.03 mm. The ovary (Fig. 23) is a broad comma-shaped organ, situated about midway between the pharynx and testis, on either side of the body. The larger part is anterior and ventral and contains many nuclei of forming ova, and there are zones of developing ova, each with larger and fewer cells until dorsally and posteriorly the oviduct is given off. The oviduct passes mediad, expanding slightly, and then posteriad and ventrad to open into the ootype. This structure is in the ventral part of the body, just anterior to the testis (Fig. 24) ; from the sides it re- ceives the vitello-vagiual canals and gives off the genito-intestinal canal. This canal after winding in a double loop opens into the intestine on the same side as the ovary. It was empty in the sectioned worm. The exteimal openings of the vaginae are situated on small prominences ventro-lateral in position, altho there is a single large opening to the exterior. The vitellaria consist of masses of follicles occupying the dor- sal and lateral areas of the body. They form a sheet of gland cells on the dorsal side of the body posterior to the testis. They are somewhat reduced along the median dorsal area in the anterior half of the worm and entirely absent onlj^ in small fields over the testis and uterus. They extend along the sides of the body and ventrally are limited by the ceca. On either side, at the level of the ootype, a common duct from the longi- tudinal collecting ducts passes ventrad and just below the cecum unites ■wath the vagina of that side to form the vitello-vaginal canal which discharges into the ootype. The uterine duct leads to the i;terus, which in each of the specimens contained a large egg. A section of the egg is shown in Figure 23. The eggs are oval, 0.15 by 0.18 mm., and in the sectioned worm the eg^ is 0.24 ram. in thickness. From the uterus a small duet passes anteriad and ventrad, opening into the genital atrium, posterior and ventral to the cirrus sac. The excretory system agrees with the general description given. 319] NORTH AMERICAS POLYSTOMIDAE—STUiXKARD 39 The descending and ascending ducts are 6 to 11^ in diameter; when emptj' their walls collapse. P. megacotijle differs from all known American forms in the large number of hooks present in the genital coronet, and in this character agrees only with P. ocellatum. The species differs from P. ocellatum, however, in the difference in size of tlie anterior sucker and pharynx as well as in the size of the caudal suckers. P. megacotyle differs from P. mkrocotyle in the number of genital hooks and in the size of the posterior suckers. P. megacotyle has a larger pharynx, larger caudal suckers, and a larger number of gental hooks than P. coronatum. POLYSTOMA MICROCOTYLE Stunkard 1916 [Figures 28 and 29] This species is described from a single specimen from the mouth of Chrysemys marginata from Creston, Iowa. The worm was stained and moiuited in toto (Fig. 28). It is 3 mm. long, and 0.78 mm. in width. The caudal disc is cordi- form, 1 mm. in width at the level of the anterior suckers, 1.07 mm. thru the middle pair, and 0.74 mm. thru the caudal pair of suckers. Each sucker is 0.28 mm. in diameter and with the exception of the longer distance between the anterior suckers, they are separated by almost regular equal distances. The distance between the anterior suckers is about four times as great as that between the posterior pair. Four larval booklets are present between the two anterior suckers, three in a row but with their hooks pointing in different directions, and the fourth some distance posterior to the others (Fig. 29). Between the posterior suckers there are three pairs of hooks : the pair of great hooks, one pair of larval hooks, and a third pair intermediate in size between the great and larval hooks. The hooks of this third pair are the same shape as the great hooks. The larval hooks are 0.017 mm. long, the gi-eat hooks are 0.116 mm. long, and the pair intermediate in size are 0.061 mm. long. In this specimen as the suckers are small the musculature of the caixdal disc shows very plainly (Fig. 29). Muscle strands from the ventral side of the body and others from the body wall pass to the bases of each of the suckers. Others pass to the outside of the dift'erent suck- ers and are inserted on the distal and intermediate zones of the suckers, serving as retractors in the operation of the organs. Many break up into smaller fibers and can not be traced. From the base of each sucker the muscles spread out in a fan shaped manner and fibers can be traced not only to the large strands from the body wall but also small fibers 40 ILLINOIS BIOLOGICAL MOXOGRAPHS [320 pass from the base of each sucker to each of the other suckers, ilan.y of the muscles brancli and ramify thru tlie tissue of the disc. The anterior sucker is 0.2 mm. long and 0.42 mm. wide ; the pharynx is 0.37 mm. long and 0.4 mm. in width. No esophagus is visible in the single toto preparation and only the anterior part of the intestine can be seen. The testis is slightly anterior to the middle of the body; it is oval, 0.36 mm. in length and 0.42 mm. in width. The sperm duct can be traced dorsally and anteriorly; cephalad of the ovary it expands into a seminal vesicle which stains deeply due to the presence of spermatozoa. The genital coronet contains thirty-two hooks, equal in size and similar in shape. The ovary is on the left side of the body, about midway between the testis and the genital pore. The oviduct arises at the median pos- terior margin and passes mediad, but the structure of the ootype could not be made out. The uterus can be distinguished at the level of the ovary on the opposite side of the bodj' aud is empty. Laterally the vaginae are visible and the vitello-vaginal canals can be traced mediad a short distance from the ceca. The vitellaria are .strongly developed, anteriorly they extend to the middle of the pharynx, and posteriorly to the caudal disc. There is a strand of follicles across the body from side to side between the pharynx and the level of the genital pore. The follicles occupy the dorsal and lateral regions of the body but anteriorly are reduced in the median area and are absent in the fields over the testis and ovary. They obscure the ceca caudal to the testis. No vitel- line ducts were seen. The excretory vesicles appear one on either side of the body dor- sally, at the level of the bifurcation of the intestine. In number of genital hooks this specimen agrees only with P. coro- iwtmn Leidy. A comparison with a type specimen of P. coronatum shows that in the latter form the pharynx and testis are much smaller and the suckers of the caudal disc are much larger. POLYSTOMA CORONATUM Leidy 1888 [Figure 27] This description was made from a single type specimen from the United States National Museum. The worm was stained and mounted in toto. Leidy (1888) says the host is the common food terrapin, and the previous year, speaking of eating terrapin, he mentions Emys ijalustris and Emys rugosa. Braun (1879-1893) lists the species from Cistudo 321] XORTH AMERICAX POLVSTOMIDAE—STUXKARD 41 Carolina. Goto (1899) in discussing the specimen described by Leidy as P. ohlongum, refers to the food terrapin as E. rugosa. Leidy gives no figure and his description states: "Polystomum coro- natum Body when elongate lanceolate. Caudal disc wider than the body, cordiform, with three pairs of bothria and with the body attached between the anterior two pairs; changeable in form to oblong, circular or quadrate ; with three pairs of minute hooks between the anterior part of bothria and with a larger pair and two smaller pairs between the last pair of bothria. Genital aperture with a circular or transverse oval coronet of thirty-two hooks of equal length. No eyes visible. Length, elongated from 4 to 6 mm ; contracting to about half the length and widening proportionately." The specimen from wliieh the present description was made (Fig. 27) is 3.15 mm. long and 0.83 mm. in width. The greatest width is at the level of the vaginae ; the body tapers rapidly anteriorly, widening again slightly at the anterior sucker. From the level of the vaginae the body gradually grows narrower posteriorly to its insertion into the caudal disc. The disc is 1.24 mm. wide at the level of the anterior suckers, 1.2 mm. thru the middle pair and 0.78 mm. thru the caudal pair of suckers. Each sucker is approximately 0.37 mm. in diameter, and constructed as previously described. There are thirty-two small di- visions in the peripheral cuticular band of the only sucker in which they could be counted. The disc bears the usual eighteen hooks ; the six larval booklets at the anterior margin of the disc are situated in a row equidistant from the anterior edge of the disc, the two lateral hooks on either side are nearer each other than the more centrally located one is to the median one of that side. Larval booklets are present in the bases of the suckers and one pair is present between the caudal suckers. Between the caudal suckers there are present also both a pair of great hooks and a third pair intermediate in size between the two. The larval booklets are 0.02 mm. in length, tlie hooks of intermediate size are 0.051 mm., and the great hooks are 0.132 mm. The anterior sucker is oval, 0.16 mm. long and 0.4 mm. wide ; the pharynx is circular in outline, 0.3 mm. in diameter. No esophagus can be seen in the toto preparation and behind the posterior margin of the testis the ceca are obscured by the vitellaria. The testis is slightly anterior to the center of the body, circular in outline, and 0.3 mm. in diameter. The vas deferens could not be distin- guished; the cirrus sac in ventral aspect is 0.19 mm. in diameter; the genital coronet contains thirty-two hooks, similar in size and shape, the shanks being sickle-shaped. The ovary is situated on the right side of the body, about its own 42 ILLINOIS BIOLOGICAL MONOGRAPHS [322 diameter anterior to the testis ; in ventral view it is circular, 0.094 mm. in diameter. The oviduct passes posteriad and mediad, and the ootype appears as a darkly stained area. The vaginae can be distinctly seen and laterad of the ceca on either side there is a large cavity communi- cating with the exterior. The uterus is empty; the folded walls of the cavity are visible on the left side of the body. The vitellaria are strongly developed. Masses of follicles occupy the dorsal and lateral regions of the body biit ventrally do not extend mediad of the ceca. Anteriorly they extend to the region of the pharynx ; there is a strand across the body just behind the pharynx and in the intercecal area anterior to the testis they are largely interrupted, permitting the structures in this region to be made out. None of the vitelline ducts are visible. The excretory vesicles are anterior to and slightly laterad of the ceca at the level of the caudal margin of the pharynx, but no ducts could be seen. POLYSTOMA HASSALLI Goto 1899 [Figures 30 to 33] This species was described by Goto (1899) from the urinary blad- der of Cinosternum pennsylvanicum from Maryland. The writer has since collected the species from other hosts and localities. A single specimen was found in the urinary bladder of Aromochehjs carinatus from Newton, Texas; five were collected from the iirinary bladder of Aromochehjs odoratus from Ealeigh, North Carolina; two from the urinar}- bladder of Cinosternum pennsylvanicum from Raleigh, N. C; and three from the urinary bladder of Chelydra serpentina from Walker, Iowa. The worms (Figs. 30, 31) vary from 1.3 to 2 mm. in length and from 0.4 to 0.65 mm. in width. The caudal disc varies in shape from hexagonal to cordiform and is of approximately the same width as the body. The suckers are 0.12 to 0.16 mm. in diameter. The eighteen hooks of the caiidal disc have the iisual arrangement and are described by Goto. HoM'ever, he reports the larval hooks as being 0.33 mm. in length and the great hooks between the caudal suckers as 0.125 mm. in length. This is evidently a typographical error, since he figured the great hooks as about four times the size of the small ones. In the pres- ent material the great hooks are the same length as stated by Goto and tlie smaller ones are 0.033 mm. in length, which agrees with the figures of Goto bj' a change of one place in the decimal point. Tlie anterior sucker is ovoid, more pointed anteriorly. It may be longer in either the anterior-posterior or lateral axis and varies in diame- ter from 0.22 to 0.33 mm. The pharynx is spherical or oval and varies 323] NORTH AMERICAN POLYSTOMIDAE—STUNKARD 43 in width from 0.1 to 0.14 mm. ; it may be longer in either axis. There is no esophagus, but in some specimens a median pocket of the intestine extends anteriad from the bifurcation to the pharynx. In others, and this is a more usual condition, lateral pockets of the intestine extend anteriad, one on either side of the pharynx (Fig. 33). The anterior sucker and pharynx are lined with cuticula; the intestine with the usual digestive epithelium. In those specimens in which the uterus contains an egg, the large size of the egg causes the ceca to be widely separated at the uterine level and they approach each other behind the uterus. In one specimen, median branches from the two ceca fuse and form a posterior connection of the diverticula (Fig. 30), and in another the two ceca are united at their ends. The testis is situated ventrally, just behind the middle of the body. It is a somewhat shapeless mass, roughly oval in outline, crosswise of the body, extending between the ceca just posterior to the uterus. The vas deferens passes anteriad, dorsal to the ovary and between it and the uterus; anterior to the uterus the sperm duct turns ventrad, enlarges to form a seminal receptacle, and then passes thru the cirrus sac, opening into the genital atrium (Fig. 32). The cirrus hooks are sixteen in number, 0.028 mm. in lengtli, straight, and with a wing like process at the middle as described by Goto. The ovary is comma shaped or ovoid in outline, situated obliquely in the body, on either the right or left side. Typically the ovary is on one side of the body and the uterus on the other, but the enormous size of the egg causes the uterus to occupy a more or less central position, crowding the ovary far to one side. The ovary is 0.058 by 0.065 mm. in the smallest and 0.085 by 0.12 mm. in the largest worms, altho the size of the ovary does not correspond precisely with the size of the worm. The oviduct arises at the dorsal median and posterior part of the ovary and after a dorsal loop it turns posteriad and ventrad to open into the ootype. Mehlis' gland is present. The genito-intestinal canal branches from the ootype and after a short winding course opens into the intestine near the ovary. From the ootype, the uterine duct passes laterally to the opposite side of the median line and then anteriorly and dorsally to open into the dorsal posterior part of the uterus. The vitellaria extend from the pharyngeal region to the anterior margin of the caudal disc; there is a row of follicles across the dorsal surface behind the pharynx but they are absent between the ceca anterior to the testis. According to Goto, "lobes not very numerous, separated from one another, mostly confined to the lateral portion of the body, but also present in the median portion behind the testis." The vaginae are ventro-lateral, midway between the anterior and posterior ends of the body. There are no vaginal prominences, the vaginal openings are 44 ILLIXOIS BIOLOGICAL MOXOGRAPHS [324 single, aud internally they unite with ducts from the longitudinal vitel- line ducts to form the viteUo-vaginal canals, as described for the other species. They do not open separately into the ootype, but the two vitello-vaginal canals open into a common reservoir from which a duct passes dorsad and discharges into the ootype (Fig. 32). In a few of the specimens the uterus is empty aud in others it contains a single large egg, the size of which varies within wide limits. The smallest eggs are 0.11 by 0.25 mm. and the largest 0.18 by 0.34 mm. The pos- terior edge of the uterus is at the level of the vaginae, and anteriorly there is a small duct from the uterus to the ventral posterior part of the genital atrium. The genital pore is in the median line, a short distance posterior to the bifurcation of the alimentary tract. The excretory pores are slightly more posteriorly situated than in the previous described species. Descending and ascending ducts occupy the characteristic positions. POLYSTOMA OBLONGUM Wright 1879 This species was described by Wright (1879) from the urinary bladder of Aromochelys odoratus. I have had no opportunity to work on material of the species and the following discussion is based on the description of Wright. According to that author P. oblongum measures up to 2.5 mm. in length and 1.5 mm. in width. The body is oblong in shape, tho capable of considerable variation. The caudal lamina is some- what narrower than the greatest width of the body and is shorter than broad. The arrangement of the suckers and hooks is similar to that in P. integerrimum; the suckers are 0.2 mm. in diameter; the large hooks are 0.15 mm. and the small hooks are 0.015 mm. in length. The mouth is on the ventral surface of the rounded anterior end. The pharyux is bowl-shaped and the intestinal ceca are without anasto- moses or branches. The description of the excretory system is very meager; concerning it he says that only the convoluted lateral stems were observed near the anterior end. The testis is situated in the posterior third of the body, the vas deferens passing dorsad and anteriad to the genital pore, which lies im- mediately behind the bifurcation of the intestine. The cirrus coronet is described as consisting of sixteen alternately large and small hooks. The free end of each is sharply curved, while the attached end is shaped like a cross the transverse piece of which is longer on one side than the other. The longer pieces measure 20/<. and the shorter ones 15ja. Doubt is expressed concerning the disposition and relations of tlie female organs. The ovary is described as situated in front of the testis on the right side of the body, but it seems probable that the organ figured 325] NORTH AMERICAN POLYSTOMIDAE-STUN KARD 45 as the " (shell gland?) "' is really the ovaiy. The lobes of the vitellaria are scattered and extend from the pliarynx to the caudal lamina or disc. It is doubtful whether Wright was correct in his statement that "The transverse duct seemed to pass inward dorsally from the intestinal ceca," since in all other known species the vitelline ducts are ventral in position. The uterus is described as containing a single large egg or embryo. The egg shell is thin and is destitute of the short stump present in that of P. integcrrimitm, but has a rather large operculum. In two cases the embryo had already escaped from the shell and moved actively within the uterine chamber. It is a Gyrodactylus-like larva, similar to that of P. integerrimum, with eye spots disposed in the same fashion. It is devoid of cilia, and movement seemed to depend entirely on the muscles and hooks of the caudal disc. The latter had a rounded outline except posteriorly where there was a square projection bearing the four small posterior hooks. The disc measured 0.114 mm. across and the twelve small anterior hooks were disposed at regular intervals on the margin of the rounded part. There was no trace of suckers. The small hooks had already attained their definitive size and form. The two large hooks were situated considerably further in from the margin than in the adult, and measured only 0.024 mm. instead of 0.15 mm. in length, which dif- ference it is stated was due to the shortness of the immersed portion, in which, however, the notch was already formed. In shape, as well as relative position and size of organs, P. oblangmn strongly resembles P. hassalli. It is significant also that both are from the urinary bladder of AromocheUjs odoratus. P. oblongum is slightly longer and broader than P. hassalli, the posterior suckers are larger and the small hooks of the disc are only about half the length of those in those in P. lutssalli. The two species agree in number of genital hooks, but in the former species the hooks are alternately large and small and with the free end sharply curved, while in P. hassalli they are straight and luiiform in size. The species in the genus Polystoma have been arranged in the form of an analytical key utilizing the more prominent or more useful diagnostic structures in separating the different forms. This key is found on the following page. 46 ILLINOIS BIOLOGICAL MONOGRAPHS [326 KEY TO THE SPECIES OP THE GENUS POLTSTOMA Uterus long, contains many eggs 2 Great hooks present on the caudal disc 3 Ceca branching P. integerrimuni Ceca not branching _ P. hulliense Great hooks not present on caudal disc P. alluaudi Uterus short, contains a single egg 7 Great hooks present on caudal disc _ 8 Genital hooks of equal length 9 Not more than sixteen genital hooks 10 Genital hooks eight in number; ectoparasitic form P. integerrimuni Genital hooks sixteen in number P. hassalli Genital hooks more than sixteen in number „ _ 13 Genital hooks thirty-two in number.... _ 14 Caudal suckers large, adjacent but not contiguous, pharynx, smaller than anterior sucker P. coronatum Caudal suckers small, widely separated, pharynx equal in size to anterior sucker P. microcotyle Genital hooks more than thirty-two in number 17 Testis simple _ ~ — 18 Caudal suckers large, overlap P. megacotyle Caudal suckers small, separated P. ocellatum Testis branched J", kachugae Genital hooks unequal in length JP. oblongum Great hooks of caudal disc reduced or absent 23 Genital hooks sixteen in number P. oriictdare Genital hooks thirty-two in number....„ _ — — P. opacum 1 6) 2 ' 5) 3 ' 4) 4 3) 5 ' 2) 6 1) 7 22) 8 21) 9 12) 10 11) 11 10) 12 9) 13 16) 14 15) 15 14) 16 13) 17 20) 18 19) 19 18) 20 17) 21 8) 22 7) 23 24) 24 23) 327] NORTH AMERICAN ASPJDOGASTRIDAE—STCNKARD ASPIDOGASTRIDAE Because of its peculiar multiloeulate adhesive apparatus, Burmeister (1856) called attention to the difference between the genus Aspidogaster and the remainder of the trematodes, and suggested a division of the Trematoda into (1) Malacobothrii for the distomes and holostomes, (2) Pectobothrii for the polystomes, and (3) Aspidobothrii for Aspido- gaster. Subsequent writers however continued to include Aspidogaster with the polystomes until Monticelli (1892) revived the classification of Burmeister, but named the three suborders into which he divided the trematodes, Heterocotylea, Aspidocotylea, and Malacocotylea. In the classification of Monticelli, the Aspidocotylea contained the single family Aspidobothridae. Poche (1907) proposed to make the name of the family agree with the rules of zoological nomenclature according to which "The name of the family is formed by adding the ending -idae to the stem of the name of its type genus. ' ' Thus the name of the family must become Aspidogastridae. The family is of special interest to students of trematode mor- phology. The form of the adhesive apparatus, with its retractile mar- ginal organs, the separation of the body into dorsal and ventral portions by a muscular partition, the sac-like alimentary tract, and the details of the genital organs are peculiar to the group. The family contains both ectoparasitic and endoparasitic species, forms with direct develop- ment and at least one species which has an intermediate host, while the hosts infested by the adult parasites include both invertebrates and vertebrates, species having been reported from molluscs, fishes, and turtles. Summaries or revisions of the group have been made by Diesing (1850, 1859), Taschenberg (1879), Hoyle (1888), Moutieelfi (1892), Braun (1879-1893), and Nickerson (1902). Only three species representing two genera of the family are known from North America, Aspidogaster conchicola von Baer 1827, Cotylaspis insignis Leidy 1856, and Cotylaspis cokeri Barker and Parsons 1914. Representatives of each of these species were available for the present study. The first two species are well known; concerning A. conchicola no further data were obtained, but a few corrections are made to former descriptions of C. insignis. 48 ILUXOIS BIOLOGICAL MONOGRAPHS [328 Cotylis iiisignc. Rieii. Rep. St. Lab. Nat. Hist. Springfield, III. Frohlich, J. A. 1789. Beschreibungen einiger neuer Eingevveidewiirmer. Naturforscher, Halle, 24:101-162, figs. 1-31. (Cited after Stiles and Hassall, 1902-1912.) 1791. B'eytrage zur Naturgeschichte der Eingeweidewiirmer. Naturforscher, Halle, 25:52-113, 17 figs. (Cited after Stiles and Hassall, 1902-1912.) 367] XORTH AMERICAS POLYSTOMIDAE—STUNKARD 87 GOLDSCHMIDT, R. 1902. Untersuchungen iiber die Eireifung, Befruchtung und Zelltheilung bei Polystomum iiitegerriiiium Rud. Zeit. f. wiss. Zool., 71 :397-44S, 3 pi. 1902a. B'emerkungen zur Entwicklungsgeschichte des Polystomum integer- rimuiit Rud. Zeit. f. wiss. Zool., 72:180-189, 11 figs. 1909. Eischale, Schalendriise und Dotterzellen der Trematoden. Zool. Anz., 34:481-498. Goto, S. 1894. Studies on the Ectoparasitic Trematodes of Japan. Jour. Coll. Sci., Imp. Univ. Japan, 8, pt. i, 300 pp., 27 pi. 1899. Notes on Some Exotic Species o£ Ectoparasitic Trematodes. Jour. Coll. Sci., Imp. Univ. Japan, 12:263-295, 2 pi. Halkin, H. 1902. Recherches sur la maturation, la fecondation et la developpement du Polystomum integcrrimum. Arch. d. Biol., 18:291-359, pi. 1014. H.\SWELL, W. A. 1907. Genito-intestinal Canal in Polyclads. Zool. Anz., 31 :643-644. HOYLE. W. E. 1888. Trematoda. Encyc. Brit. 9 ed., 23 :535-540. Iji.\i.\, I. 1884. Ueber den Zusammenhang des Eileiters mit dem Verdauungscanal bei gewissen Polystomeen. Zool. Anz., 7 :63S-639. Jacerskiold, L. a. 1899. Ueber den Bau von Macraspis clegans. Ofvers. Vet-Akad. Fordh., 3:197-214, I pl- Johnston, S. J. 1912. On Some Trematode Parasites of .■Xustralian Frogs. Proc. Linn. Soc. X. S. Wales, 37 :285-362, pi. 14-43. 1914. Trematode Parasites and the Relationships and Distribution of their Hosts. Rep. Austral. Ass. Adv. Sci., 1914:272-278. Kelly, H. M. 1899. A Statistical Study of the Parasites of the Unionidae. Bull. III. St. Lab. Nat. Hist., 5 :399-4i8. KOFOID, C. A. 1899. On the Specific Identity of Cotylaspis iitsigitis Leidy and P!atyasl>is aitadontae Osborn. Zool. Bull. Boston, 2:179-186. KuHL, H., und Hasselt, J. C. van 1822. Polystoma midac. Isis, 1822:113-115. (Cited after Braun, 1879-1893.) KUHN, J. 1829. Description d'un nouvel epizoaire du genre Polystomum qui se trouve sur Ics branchies de la petite roussette (Squalus catutus) suivie de quelques observations sur !e Distomum mcgastomum Icforis. Ann. d. sc. d'obs., 2 : 460-465. Leidy, J. 1851. Helminthological Contributions II. Proc. Acad. Nat. Sci. Phila., 5 :224-227. 88 . \ORTH AMERICAN PARAMPHISTOMIDAE—STUNKARD [368 1857. Observations on Entozoa Found in the Xaiades. Proc. Acad. Xat. Sci. Phila., 9:18. 1858. Contributions to Helminthology. Proc. Acad. Nat. Sci. Phila., 10:110-112. 1888. Entozoa of the Terrapin. Proc. Acad. Nat. Sci. Phila., 40:128. Linton, E. 1905. Parasites of Fishes of Beaufort, North Carolina. Bull. Bur. Fish., 24:321-428, 34 pi. Looss, A. 1885. Beitrage zur Kenntnis der Trematoden. Zeit. f. wiss. Zool., 41 :390- 446, I pi. 1892. Ueber Amfhistomuiii subclavatum Rud. und seine Entwicklung. Festschr. Leuckart, p. 147-168, pi. 19-20. 1893. 1st der Laurer's Canal der Trematoden eine Vagina? Centr. f. Bakt. u. Par., 13:808-819. 1896. Recherches sur la Faune parasitaire de I'Egjpt. Premiere partie, Mem. de ITnst. egypt, 3:1-252, 16 pi. 1899. Weitere Beitrage zur Kenntnis der Trematoden-Fauna Aegyptens. Zool. Jahrb., Syst., 12:521-784. 1902. Ueber neue und bekannte Trematoden aus Seeschildkroten. Zool. Jahrb., Syst., 16:411-894, pi. 21-32. 1912. Ueber den Bau einiger auscheinend seltener Treraatoden-Arten. Zool. Jahrb., Suppl., 15:323-366, 3 pi. LiJHE, M. 1909. Parasitische Plattwiirmer. I, Trematodes. Die Siisswasserfauna Deutschlands, 17:1-217, 188 figs. MacC.^llum, G. a. 1913. Fertilization and Egg-Laying in Microcoiyle stenotomi- Science, n. s., 37:340-341- MacCallum, G. a., and MacCallum, W. G. 1913. On Aspidogaster ringens and A. kemostoma n. sp. Zool. Jahrb., Syst., 34:245-256, 4 figs. MacCallum, W. G. 1905. On Two New Amphistome Parasites of Sumatran Fishes. Zool. Jahrb., Syst., 22:667-678, 2 figs. Mace, E. 1880. Des Trematodes parasites des Grenouilles. Bull. Soc. d'etud. scien. Finistere, Morlaix, 31 pp., 4 pi. (Cited after B'raun, 1879-1893.) MONTICELLI. F. S. 1892. Cotylogaster michaclis n.g., n. sp., e revisione degli Aspidobothridae. Festschr. Leuckart, p. 166-214, 2 pi. 1903. Per una nouva classificatione degli "Heterocotylea". Monit. zool. Ital., 14:334-337- NiCKERSON, W. S. 1895. On Stichocotyle nephropis Cunningham, a Parasite of the American Lobster. Zool. Jahrb., Anat., 8:447-480, 3 pi. 369] XORTH AMERICAS POLYSTOMIDAE—STUSKARD 89 1902. Cotologaster occidcntalis n. sp. and a Revision of the Family Aspido- bothridae. Zool. Johrb., Syst, 15 :S97-624, 2 pi. Odhner, T. 1898. Uber die geschlechtsreife Form von Stichocotyle nepliropis Cunning- ham. Zool. Anz., 21 :509-5i3. 1911. Zum natiirlichen System der digenen Trematoden, I. Zool. .*\nz., 37: 181-191. 1911a. Zum natiirlichen System der digenen Trematoden, IV. Zool. .'\nz., 38:513-531- 1912. Die Homologien der weiblichen Genitalwege bei den Trematoden und Cestoden. Nebst Bemerkungen zum natiirlichen System der monogenen Trematoden. Zool. Anz., 39:337-351. 1913. Noch einmal die Homologien der weiblichen Genitalwege der mono- genen Trematoden. Zool. Anz., 41 :558-SS9. OSBORN, H. L. 1898. Observations on the Anatomy of a Species of Platyaspis Found Para- sitic on the Unionidae of Lake Chautauqua. Zool. Bull., 2 :5S-67, 6 figs. 1904. On the Habits and Structure of Cotylasfis iiisignis Leidy, from Lake Chautauqua, N. Y. Zool. Jahrb., Anat., 21:201-243, 3 pi. Otto, H. R. 1896. Beitrage zur Anatomie und Histologic der Amphistomideii. Dtsch. Zeit. Thiermed., 22:85-141, 30 figs. PocHE, F. 1907. Einige Bemerkungen zur Nomenclature der Trematoden. Zool. .\nz., 31 :i24-i26. POISIER, J. 1886. Trematodes nouveau.x ou peu connus. Bull. Soc. Philom. Paris, 7 ser., 10 :20-40, 4 pi. Pratt, H. S. 1900. Synopses of North American Invertebrates, XII. The Trematodes. Part I. The Heterocotylea or Monogenetic Forms. Amer. Nat., 34 :645- 662, 50 figs- 1908. Parallel Development in Trematodes. Science, u. s., 27 :489. RUDOLPHI, C. A. 1801. Bebachtungen iiber die Eingeweidewiirmer. Arch. f. Zool. u. Zootom., 2:1-65. 1809. Entozoorum sive vermium intestinalium historia naturalis. Amste- laedami, 1809. 1819. Entozoorum synopsis. Berolini, 1819. St. Remy, G. 1891. Synopsis des Tremotodes monogeneses Rev. Biol. Nord. France, 4:1-92, 2 pi. i8g8. Complement du synopsis des Trematodes monogeneses. Arch. d. Paras., I :S2i-S7i, 6 figs. Stafford, J. 1896. Anatomical structure of Aspidogaster conchicola. Zool. Jahrb., .Anat., 9:477-542, 4 pl. 1900. Some Undescribed Trematodes. Zool. Jahrb., Syst., 13:399-414, i pl. 90 KORTH AMERICAN POLYSTOMIDAE—STUNKARD [370 1905. Trematodes from Canadian Vertebrates. Zool. Anz., 28:681-694. Stewart, F. H. 1914. The Anatomy of Polystomum kachugae n. sp. with Notes on the Genus Polystomum. Rec. Ind. Mus., 10:195-205, 4 pi. Stieda, L. 1870. Uebcr den Ban des Polystomum intcgerrimum. Arch. f. Anat. Phys. u. Med., 5:660-678, I pi. Stiles, Ch. W., and Goldberger, J. igio. A Study of the Anatomy of Wahonius watsoni of Man and of Nine- teen Allied Trematode Worms of the Superfamily Paramphistomoidea. Bull. Hyg. Lab., No. 60:1-264, 205 figs. Stiles, Ch. W., and Hassall, A. 1902-1912. Index-Catalogue of Medical and Veterinary Zoology. Bull. Bur. Animal Ind., No. 39. 1908. Index-Catalogue of Medical and Veterinary Zoology. Bull. Hyg. Lab., No. 37:89- Stunkard, H. W. 1916. On the Anatomy and Relationships of Some North American Trema- todes. Jour. Par., 3:21-27. Taschenberg, O. E. 1879. Zur Systeraatik der Monogenetischen Tematoden. Zeit, gesammt Naturwisscnsch., 52:232-265. Treutler, F. A. 1793- Observationes pathologico-anatomicae, auctariam ad helminthologiam humani corporis continentes. Diss, in praes. Ch. F. Ludwig. Lips. (Cited after Braun, 1879-1893.) VOELTZKOW, A. 1888. Aspidogaster conchicola. Arb. zool.-zootom. Inst. Wurzb., 8:249-292, 5 Pl. Walter, E. 1893. Untersuchungen uber den Bau der Trematoden. Zeit. f. wiss. Zool., 56:189-243, 3 pl. Willemoes-Suhm, R. von 1872. Zur Naturgeschichte des Polystomum integerrimum and Polystomum ocellatum. Zeit. f. wiss. Zool., 22 :29-39, i pl. Wright, R. R. 1879. Contributions to American Helminthology. Jour. Proc. Canad. Inst, N. S., I :i-23, 2 pl. Wright, R. R., and Macallum, A. B. 1887. Sphyranura osleri. Jour. Morph., I :i-48, i pi. Zeder, J. G. H. 1800. Erster Nachtrag zur Naturgeschichte der Eingeweidewiirmer von J. A. G. Goeze. Leipzig, 1800. Zeller, E. 1872. Untersuchungen uber die Entwicklung und Bau des Polystomum in- tegerrimum Rud. Zeit. f. wiss. Zool., 22:1-28, 2 pl. 1876. Weiterer Beitrag zur Kenntniss der Polystomen. Zeit. f. wiss. Zool., 27:238-275, 2 pl. 3711 XORTH AMERICAN POLYSTOMIDAE—STUNKARD EXPLANATION OF PLATES All figures except those of reconstruction were drawn with the aid of a camera lucida and were made from permanent mounts. Abbreviat ;ions used a acetabulum nc nerve commissure b esophageal bulb ovary cm circular muscles oc eye spot cs cirrus sac od oviduct e esophagus om oblique muscles ed excretory duct 00 ootype ep excretory pore op oral evagination 9P genital pore OS oral sucker gc genito-intestinal canal ov egg h small booklets p postate gland hd hermaphroditic duct ph pharynx i intestine sp septum I Laurer's canal sv seminal vesicle Im longitudinal muscles t testis Is lymph sinus u uterus It limiting membrane ud uterine duct m mouth V vitellaria iiid median dorsal lymph canal vd vas deferens mg Mehlis' gland vg vagina mo marginal organ vl vitelline duct mt metraterm w vitello-vaginal canal mv median ventral lymph canal 373] NORTH AMERICAN POLYSTOMIDAE-STUNKARD 93 PLATE I 94 ILLINOIS BIOLOGICAL MONOGRAPHS [374 EXPLANATION OF PLATE POLYSTOMA ORBICULARE Fig. I. Entire specimen, extended, ventral view. X 35. Fig. 2. Hook from genital coronet. X 225. Fig. 3. Reconstruction of genital apparatus from frontal sections. X I3S. Fig. 4. Sagittal section thru caudal disc. X 87. Fig. 5. Frontal section thru caudal disc. X 7Z- Fig. 6. Sagittal section thru oral sucker and pharynx. X 140. ILLINOIS BIOLOGICAL MOXOGRAPHS l-OLUME 3 STUXKARD NORTH AMERICAN POLYSTOMIDAK PLATE I 375] NORTH AMERICAN POLYSTOMWAE—STUNKARD 95 PLATE II 96 ILLINOIS BIOLOGICAL MOXOGRAPIIS [376 EXPLANATION OF PLATE POLYSTOMA ORBICULARE Fig. 7. Frontal section. X 35. Fig. 8. Frontal section of ootype and beginning of uterine duct. X 185. Fig. 9. Frontal section of ootype and end of right vitello-vaginal canal, five sections ventral to Figure 8. X 185. Fig. 10. Frontal section, oiitype region of same specimen as Figures 8 and 9, showing ovary, uterus, oviduct, uterine duct, genito-intestinal canal and vas deferens. X 140. Frontal section showing vitellaria and origin of vitelline ducts with granular secretion in the cells and duct. X 87. Frontal section thru cirrus sac at the juncture of the shanks and roots of the genital hooks, showing the genital papillae cut across, and a sec- tion of the duct from the uterus at the bottom of the figure. X 250. Reconstruction of male genital apparatus from sagittal sections. X 140. Frontal section thru uterus showing embryo in stage of a morula-like mass of cells. X 700. Fig. II. Fig. 12. Fig. 13- Fig. 14- ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 OK OU STUXKAKD XORTll AMKRICAX POI.VSTOMIDA1-: PLATP. II XORTH AMERICAX POLrSTOMID.iE—STUXK.-lRD PLATE III 98 ILLIXOIS BIOLOGICAL MONOGRAPHS [378 EXPLANATION OF PLATE POLYSTOMA OPACUM Fig. 15. Entire specimen, extended, ventral view. X 20. Fig. 16. Reconstruction of genital apparatus from toto preparation and cross sections. X 50. Fig. 17. Hook from genital coronet. X S50. Fig. 18. Frontal section thru the anterior sucker and pharynx, showing in section nerve commissures and vitellaria. X 60. Fig. 19. Cross section of body thru uterus and cirrus sac. X 60. Fig. 20. Cross section of body thru the testis. X 60. Fig. 21. Cross section thru the anterior pair of bothria. X 60. ILLIXOIS BIOLOGICAL MOXOGRAPIIS STUXKARD NORTH A^rERICAX POLVSTOMIDAE PLATE III 379] .\ORTH AMERICAS POLYSTOMIDAE—STUXKARD 99 PLATE IV 100 ILLINOIS BIOLOGICAL MONOGRAPHS [380 EXPLANATION OF PLATE POLYSTOMA MEGACOTYLE Fig. 22. Entire specimen, ventral view. X 27. Fig. 22- Cross section of body thru ovary and uterus. X 60. Fig. 24. Cross section of body thru vaginae and anterior part of the testis. X 60. Fig. 25. Cross section thru the pharynx near the posterior end. X 85. Fig. 26. Cross section of seminal vesicle and cirrus sac. X 140. POLYSTOMA CORONATUM Fig. 27. Entire specimen, ventral view. X 27. ILLISOIS niOLOCICAL MOXOCRAPHS 2:i VOLUME 3 26 K TV* STLXKARO XORTll AMF-.RICAX POLVSTOMIDAE PLATE IV 381] NORTH AMERICAN POLYSTOMIDAE—STUNKARD 101 PLATE V mZ ILLINOIS BIOLOGICAL MONOGRAPHS [382 EXPLAXATIOX OF PLATE POLYSTOMA MICROCOTVLE Fig. 23. Entire specimen, ventral view. X 27. Fig. 29. Ventral view of caudal disc, showing arrangement of musculature and hooks. X 43- Poi.YSTOMA HASSALLI Fig. 30. Entire specimen, ventral view, ceca connected posteriorly. X 45- Fig. 31. Entire specimen, ventral view, in whicli there is no posterior connexion between the ceca. X 40. , Fig. Z2. Reconstruction of genital apparatus from frontal sections. X 135- Fig- 33- Frontal section thru the dorsal part of the uterus, showing oral sucker, pharynx, nerve commissures, intestine, excretory vesicles and ducts, vitellaria and smaller tubes of the ootype region. X 60. ILLIXOIS BIOLOGICAL MOXOCR.IPUS VOLUME 3 STL'XKAKD XUKTli A.MF-.RICAX POLVSIdMIDAK PLATE V 38o] XORTH AMERICAN POLYSTOMWAE—STUKKARD 103 PLATE VI 104 ILLINOIS BIOLOGICAL MONOGRAPHS 1384 EXPLAXATIOX OF PLATE Fig. 34- Fig. 35. Fig. 36. Fig. 37- Fig. 38. Fig. 39. Fig. 40. Fig. 41. Fig. 4^. Fig. 43. Fig. 44. Fig. 45- Suckers and Hooks of Various Species of Polystomes Polystoiim orbiculurc. bothriuiii from caudal disc. X 140. Polystoiiia orbicularc, frontal section thru botlirinm. X 140. PolystoiiH} obknlare, optical section of bothrinni showing cuticnlar framework. X 140. ' Polystonta ol^acttiii, hook from base of sucker. X 165. Polystoina oparnn:, hook from anterior margin of caudal disc. X 165. Polysloina tr.iirocotylc, hooks of posterior margin of disc. X 165. Polystoina o/'acii:}:. hooks of posterior margin of disc. X 165. Polystoina ir.cgacotylc, hooks of posterior margin of disc. X 165. Polystoina coronatuiii, hooks of posterior margin of disc. X 165. Polystoina orbicularc, hook from base of sucker. X 165. Polystoina orbicularc. frontal section thru a sucker illustrating the method of operation; the e.xtcrnal zones are retracted with the resulting protrusion of the basal part. X 140. Polystonia intcgerriiiiuiu. frontal section thru a sucker sliowing type of cuticnlar framework. Compare with text and types illustrated in other figures. X 100. lE-LIXOIS BIOLOGICAL MOXOCRAPHS rOLLME 3 f^^^'^ y STUXKARD XORTII AMF.RICAX POLVSTOMIDAE PLATE VI 385] NORTH AMERICAX ASPIDOCASTRIDAE—STVNKARD 105 PLATE VII 106 ILLINOIS BIOLOGICAL MONOGRAPHS [386 EXPLANATION OF PLATE COTYLASPIS COKERI l^ntire specimen, extended, dorsal view. X 40. Ventral view of entire specimen showing position of marginal organs and divisions of the adhesive disc. X 40. Reconstruction of reproductive organs from frontal sections. X 80. Cross section of body at the le\el of the ovary showing the ovary, uterus, seminal vesicle, intestine, excretory ducts, and a follicle of the vitellaria. X 87, Diagrammatic representation of the excretory system from a li\iii}; specimen, dorsal view. X 40. ObIii|Ue section of body just posterior to tlie genital pores, showing in section the mouth funnel, pharyn.x, cirrus sac, uterus, septum iiiid adhesive disc. X 87. Pig. 5J. Entire specimen, contracted, dorsal view. X 40. Fig. 46. Fig. 4;. Fig. 48. Fig. 49. Fig. 50- Fig. 51- ILLISOIS BIOLOGICAL ilOXOGRAPHS 46 VOLUME 3 STUXKARD NORTH A.Ml- RICAX ASriDOGASTRIDAK PLATE VII 387] NORTH AMERICAN ASFIDOGASTRIDAE— STINKARD 107 PLATE VIII 108 ILLIXOIS BIOLOGICAL MOKOCRAPllS [388 EXPLANATION OF PLATE COTYLASPIS COKERI (EXCEPT FiCURE 56) Fig. 53. Sagittal section thru the anterior end of body showing musculature, di- gestive and reproductive organs. X 200. Fig. 54. Frontal section thru the openings of the genital pores. X 85. Fig. 55. Section thru a marginal organ ; a muscle fiber is seen at the left of the figure and on the other side a nerve fibril passes to the inner end of the thick walled part of the canal. In this section the canal is ci!t across and can not be traced from the bulb to the e.xterior. X s8o- Fig. 56. Section thru a marginal organ in Cotylaspis insignis. X 580. Fig. 57. I'Vontal section thru tlic adhesive disc showing arrangement of muscula- ture. X 93- Fig. 58. Section thru the anterior part of the forebody showing the base of the mouth funnel, anterior part of the pharynx, nerve commissure and eye spots. X 800. ILLIXOIS BIOLOGICAL MOSOGRAPHS VOLUMn 3 ^ t I STUXKARI) NORTH AM F.UICAX ASPIDOC.\STI-:iJ).\i-. PLAI i:\lll 389] NORTH AMERIC.-IX PARAMPHISTOMIDAE—STVXKARD 109 PLATE IX no ILLIXOIS BIOLOGICAL MONOGRAPHS [390 EXPLANATION OF PLATE Alassostoma magnum Fig. 59- Entire specimen, ventral view. X 9. Fig. 6o. Cross section thru the genital pore showing the terminal parts of the cirrus sac and uterus, the hemaphroditic duct, genital sinus, four layers of muscles in the body wall and the muscle lamellae of the esophageal bulb. X27. Fig. 61. Diagrammatic representation of female genital apparatus reconstructed from cross sections. X40. Fig. 62. Section of the wall of the intestine. X36o- Fig. 63. Cross section thru the oral sucker and the oral evaginations. X 4°- Fig. 64. Cross section of body at the level of the ovary showing in section the ovary, uterus, Laurer's canal, the ceca, vitellaria, lymph spaces and excretory ducts. X 16. Fig. 65. Cross section thru the oral sucker showing arrangement of muscle fibers and position of the nuclear zone. X 35- ILLIXOIS BIOLOGICAL MOXOCRAPHS y GLUME 3 60 .,, O K STUXKARD XOKTII AMF.RICAX rARAMPIIISTO.MIDAE PLATE IX 391] NORTH AMERICAX PAR.UIPHISTOMIDAE—STUSKARD 111 PLATE X 112 ILLIXOIS BIOLOGICAL MOXOGRAPHS [392 EXPLANATION OF PLATE Alassostoma parvum Fig. 66. Entire specimen, ventral view. X 27. Fig. 67. Cross section of body posterior to the ovary showing coils of the excretory ducts. X 70. Fig. 68. Cross section thru the posterior part of the acetabulum showing lymph spaces around the sucker. X 70. Fig. 69. Cross section a short distance posterior to tlie genital pore showing in section, the uterus, the cirrus sac, and above the latter organ three loops of the seminal vesicle. X 70. Fig. 70. Cross section of esophageal bulb with clusters of surrounding cells. X 70. Fig. 71. Cross section of body thru the genital pore showing hermaphroditic duct, cirrus sac, lymph spaces and the character of the parenchyma. X 9°. ILLIXOIS BIOLOGICAL MOXOGRAPHS 67 VOLUME 3 STUXKARD XORTll A.MliRICAX PARA.MPHISTO.MIDAR PLATE X 393] XORTH AMERICAX PARAMPHISTOMIDAE—STUXKARD 113 PLATE XI 114 ILLIXOIS BIOLOGICAL MOXOGRAPHS [39+ Fig. -2. Fig. 73- Fig. 74- Fig. 75- Fig. 76. Fig. '/■ Fig. 78. EXPLANATION OF PL.\TE Zygocotyle ceratos.\ Entire specimen, ventral view. X n- Cross section of esophageal bulb, showing the arrangement of the muscle fibers. X 45- Compare with Figures 60 and 70. Cross section of body thru the origin of the oral evaginations. X 45. Sagittal section thru the anterior part of the bodj- showing oral sucker, an oral evagination and the anterior part of the esophagus. X 45- Sagittal section of posterior part of body thru one side of the acetabulum. X27. Sagittal section of the posterior part of the body near the median line, showing the ovary, eggs in the uterus, Laurer"s canal, and the shape of the acetabulum. X 27. Sagittal section thru the body one section at the side of the genital pores showing the folded wall of the uterus and the ejaeulatory duct which in this species is without a cirrus sac. X 136. Representation of the sagittal section thru the openings of Laurer"s canal and the e.xcretory vesicle. X 90- ILLIXOIS BIOLOGICAL MOXOCR.IPHS STUXKARI) XORTll AMERICAN PARAMrillSTOMIDAK PLATE XI ILLINOIS BIOLOGICAL MONOGRAPHS Vol. in ApnU917 No. 4 Editorial Committee Stephen Alfred Forbes William Trelease Henry Baldwin Ward Published under the Auspices of the Graduate School bh THE University of Illinois Copyright, 1917 By THE University of Illinois Distributed June 30, 1917 COLOR AND COLOR-PATTERN MECHANISM OF TIGER BEETLES WITH TWENTY-NINE BLACK AND THREE COLORED PLATES VICTOR E. SHELFORD TABLE OF CONTENTS PAGE Introduction 5 Materials and methods - 6 Analysis of Color Patterns ~ 13 Color Patterns and Elytral Structures 13 The Color Pattern Plan 19 Color Pattern and Pigment Development 21 Experimental Modification of Patterns 38 Geographic Variation of Patterns - 40 Colors of Tiger Beetles 46 Causes of Colors - ■■ 46 Ontogeny of Color 47 Relation of Ontogenetic Stages to Geographic Races 49 Geographic Variation in Color 52 Experimental Modification of Color _ 55 Relation of Colors and Color Patterns to Climate 56 Geographic Center of the Group on the Basis of Patterns 57 General Discussion .- - 58 Pattern Tendencies - 58 Bearing of the Color Pattern Mechanism on Orthogenesis 63 Bearing of the Pattern ^Mechanism on the Biogenetic Law 65 Summary of Conclusions „ -- 66 Patterns ..._ - 66 Color - - 67 Geography 67 Bibliography .- 68 Explanation of Plates "i 399] COLORS OF TIGER BEETLES—SHELF ORD INTRODUCTION In the analysis of characters made the basis of studies of variation, orthogeuetie trends, experimental modification and heredity, noteworthy advantages are associated with the study of large groups of species in which divergence and modification have proceeded in various direc- tions. The material should be plastic so that the laws governing response in characters can be determined. The ontogeny of the char- acters should be of such a character as to show the general ground plan of the system and its relation to the existing adult characters and their variations. It is further desirable to be able to breed the organ- ism, segregate pure lines, and cross various species. There is a strong tendenej' of late years to regard the breeding and the breeding results as superior to the other attempts at character analj'sis. This has pro- ceeded to such an extent without adequate physiological analysis that one writer (Riddle, 1909) designated the method of cross breeding "the mixing of unknowns". The primary object of this paper is to show the nature of the color and color-pattern mechanism of the elytron. In the matter of qualifications of material the tiger beetles are admirably adapted to all the needs enumerated above, but since one year at least and normally two are necessary for a generation, only a few single generations have been bred. For this reason the idea of breeding them was abandoned. It is also a purpose of this paper to- show that breeding is not the only method by which adequate analysis can be reached, i.e., unless the laws governing heredity are a system entirely a part from those governing the modification of parts during ontogeny and the normal course of A'ariation, which seems to be the- tacit assumption of various students of heredity in the not too distant past. I shall indicate further that orthogenetic tendencies, if directive tendencies are to be so named, are numerous and in a large series of species present a confusing set of groups wliieh are excessively com- plicated and reduced to any simple system, as claimed bj^ Eimer and von Linden for Lepidoptera or for a limited number of species by Wlntman, with difficulty. Still, large tendencies with numerous minor ones within them may be detected. It will be shown that the laws governing the modification of patterns apply alike to general, probably liereditary tc.'^deiicies a;ii! det:;il''d r,:.spc:t y.i under e.\j*:'rijii.':iiKl co;)- 6 ILLINOIS BIOLOGICAL MONOGRAPHS [400 ditions. It will be shown that biogenetic law must be applied with, caution and is not of such broad application as is held iu some quar- ters, being inapplicable to various characters altogether. The brilliant colors of the group are due to physical phenomena determined by Professor Llichelson, and leave no place for the bioge- netic law iu connection with the development of color during ontogeuj'. It will be shown further that color is closely correlated with general physiological condition and is modifiable by conditions which affect general metabolism. The results here pi'esented are based on several years of observation. In 1903 the writer undertook a study of, variation of the tiger beetles. The work here presented is the outgrowth of this beginning, and indeed includes some small portions regarding color patterns that were written in that year. The work has been prolonged for many reasons, but chief of these was the' very large number of species in the group and the fact that an adequate understanding of the mate- rial could not be attained without consulting many large collections. Further, the experimental results obtained in 1906 demanded a first- hand study of the variations of the species concerned and their natu- ral habitats. The accumulation of material and data was not com- pleted until 1911. Some of this had to be studied, drawings made, etc., which with numerous other duties and enterprises iiuder way made necessary much time to put it into the present form. A family with i^pwards of 1300 species of which more than 600 are in one genus and with characters which can be studied and analyzed, appeared to afford material which was sufficiently promis- ing to justify delay. In the fourteen years that have elapsed since the problem was first undertaken at the suggestion of Dr. C. B. Dav- enport, the attention of biologists has shifted from variation, which was then the chief topic of interest, to experimental modification of characters, and finally to the methods of modern genetics. Various men have made numerous suggestions regarding the work, but in its final preparation the writer has been able to use only a few of them in a general way, and an attempt is made to present tlie facts and conclusions growing out of the material as simply as possible. MATERI.\LS AND METHODS The material which has been used as the basis of this work has consisted of collections in the family Cicindelidae of the world, exten- sive collections of several North American species, repeated year-to-year collections of a few species in Illinois and Indiana, series of observations on the ontogenv of color in a small number of North American species. 401] COLORS OF TIGER BEETLES— SHELFORD 7 and experimental modification of a number of species which has assisted in the analysis of the color patterns. The collections studied have covered most of the species of the family, which is divided into several tribes by W. Horn in the Genera Insectorum (1915). The subfamilies herein named were in part given as families in the Systematischer Index of the same author (1906), in which he presented a preliminary list of the species which he later published in the Genera Insectorum. Accordingly, in my previous papers on the subject (1906, 1908, 1912. 1914, 1915) the "Index" -was followed almost entirely in the matter of nomenclature and order of arrange- ment. The groups represented in the family as outlined in Genera Insect- orum are as follows : Subfamily Ctenostomini (tree dwellei's) Number of species Pogonostoma; Madagascar 32 Ctenostoraa ; tropical America 45 Subfamily Collyrini (nearly all tree dwellers) Tricondyla; India " 27 Collyris; Oriental region __ 104 Subfamily Mantichorini (ground dwellers) IMantiea ; Africa _ 1 Mantichora ; Africa 5 Subfamily Megaeephalini (ground dwellers) Platychila ; South Africa.... 1 Pycnochila ; South America 1 Amblyehila; Western U. S. A. 2 Omus; Western U. S. A 4 Aniaria : Northeastern South America _ 1 Megacephala ; southern U. S. to Argentine, Africa, Arabia, Persia, Australia ......_ 68 Oxyehila ; middle America 25 Pseudoxychila : Andes. Costa Rica to Bolivia 1 Chiloxia ; Andes of Ecuador to Bolivia 1 Eucallia ; Andes of Ecuador and Columbia 1 Subfamily Cicindelini ; ground dwellers Dromiea ; southern half of Africa 82 Prothyma ; Africa, Madagascar, Asia, and the Malay Archipelago _ 50 Dilatotarsa ; Malay Archipeligo 1 Caledonomorpha ; New Guinea 1 Di.stipsidei'a ; Australia and New Guinea 8 Caledonica ; New Guinea 9 8 ILLIXOIS BIOLOGICAL MONOGRAPHS [402 Nickerlea ; Australia _ 2 Rhysopleura ; Australia 1 Euprosopus; Brazil 2 Langea; Peru 1 Iresia; coutinental tropical America 8 Tlierates ; ilalay Archipelago 33 Odontoehila; South America, Malay Pen. and Islands 75 Prepusa ; South America) 3 Oxygonia ; South America 15 Opistheneentrus : Brazil 1 Cieindela ; world-wide distribution 686 Eurymorpha; Africa _ _ 1 Apteroessa; India 1 1299 The group contains some 35 genera and upwards of 1300 species and subspecies. In the figures above the subspecies of Cieindela number- ed in Roman in Horn's Genera Insectontiu list, -which number 55, are included, but subspecies numbering more than 8 in llegacephala alone, and several in other genera, are not included. There are very few of these 1300 races which the writer has not seen in some one of the particularlj^ numerotis and complete collections studied. Those studied quite completely are : British Museum of Natural History ; Hope Collection, Oxford University ; Cambridge University ; Private Collection of Mr. Basil G. Nevinson, London ; Private Collection of Dr. Walther Horn, Berlin; Zoologisches Museum, Berlin; Private Collection of Doctor Gestro, Genoa ; Jardin des Plantes, Paris ; Museum of Comparative Zoology, Cambridge, ^Massachusetts ; United States National Museum ; Philadelphia Academy of Science : American Museiuu of Natural History, New York ; and the University of Chicago collection including an old collection once the property of John Akhurst, Brook- lyn, several purchases from Hermann Rolle of Berlin, and the material secured by exchange for other species in the Akhurst collection, and material purchased and collected for the writer by the University, and specimens collected on the excursions supported by the University. In addition to this the writer secured a collection of exotic material from Mr. John D. Sherman in exchange for Dytiscidae and numerous speci- mens by exchange and gift from numerous American and foreign collect- ors. Of the few species not seen several are represented in figures which show the color patterns. Many of the drawings presented are from the collections in ques- tion ard arp appropriately designated in the groups of figures in th" 403] COLORS OF TIGER BEETLES— SHELFORD succeeding pages. The meaning of the designations is as follows: B, British Museum ; C, Cambridge University ; D, Berlin ; G, Gestro ; H, W. Horn, Berlin ; M, U. S. National LIuseum ; N, Xeviusou ; 0, Oxford University; P, Paris; S, Shelford; U, University of Chicago. While none of the patterns of the genera other than Ciciudela are of a type differing from the general plan of the Ciciudela, patterns are very often wanting or veiy simple, such as the simple cross bands in Collyris. In course of the examination of the several collections named, a great abuudance of variation has beeu noted in some of the commoner representatives of the groups, not only of Ciciudela but others also. The taxonomic arrangement of Ciciudela by Doctor Horn in the Genera Insectorum is especially fortunate. He has arranged the species into a niimber of groups on the basis of the distribution of hairs on the head, thorax, abdomen, tarsi, labrum, and of other structural characters, but without reference to color patterns. He gives 174 groups apparently not duplicated in the different regions and 16 represented in more than one zoogeographic region by the same or closely related species. These 174 groups are distributed as follows : Ethiopian region, 34 ; Oriental , region, 48 ; Australian region, 22 ; Palearotic region which he extends to include China, 20; Nearctie region, 24; Neotropical region, 26. The groups foimd in more than one region and which are counted in the one with most species, are as follows : Table I Showing the Number of Species in Regions by Groups as Designated by a Common Species singularis Chd melancholica Fab doncgaleiisis Chd iiilotica Dj grrmanica L - foveolata Schm laetescripta Mtsch 10 guttata Fab striolata Ulig discrete Schm seiniciiicta Br brcfispoiiosKsW'. Horn carthagena Dej argentata Fabr irifasciafa Fabr macrocnema Chd 10 ILLINOIS BIOLOGICAL MOXOGRAPHS [404 111 the above list species occurring in two are counted in botli. So far as practicable these pilosity groiii3S have been considered in working up, arranging, and discussing the patterns. Considerable change has been made in the nomenclature and arrange- ment of species in the Genera Inseetorum as compared with Doctor Horn's Index. The paper had progressed so far with the Index as a tasis that it was thought not to be practicable to change it to agree with the newer work. The extensive collection of North American species belong to the first group in Horn's series for the Nearctic region. This group includes tranqucharica and will be referred to as tlie Tnniquebarica Group. These are cliaracterized as follows: The four anterior trochanters have fixed hairs, cheeks naked, or with isolated hairs, clij^eus often hairy. Frons mth discoidal or supraorbital hairs ; median portion of the frons never proportionately supplied with more or less short, close Ij'ing, do^vnward directed hairs ; frons never hairy above the antenual insertion. The disc of the middle frons is often hollowed out or sharply separated from the fore frons by its steepness. The first antennal segment is often thickly covered with outstanding hairs. The pronotum has at least rudiment- ary hairs, often circumdiscally and discally hairy ; hairs often long and fine and never decumbent except when very numerous ; free anterior and posterior border of the pronotum not hairy. The prosternum is always naked. The lateral portion of the breast is always thickly covered with hairs. The hind border of the femur and sometimes the foreborder also covered with fine short decumbent hairs ; hook-formed hairs never pres- ent; hairs on the hip and superorbital border most numerous. This group stands in close relation to the European group to which campestris belongs. The main group includes formosa,* venusta, limbata, purpurea* ancosisconensis, duodecemguttata,* hirticollis* latesignata, traitquchar- ica* tenuicincta, bellissima, longilahris, eureka, oregoiia, senilis, iviUi- stoni, fulgida, pidchra, pimeriana, scutellaris* In addition to this, collections were made of C. sexguttata which stands in a group by itself. Those starred were studied especially. Collections of these species representing complete catches were supplied by C. S. Brimley, E. G. Smyth, C. A. Frost, L. H. Joutel, Rev. J. C. Varren, and Dr. C. F. Adams. Collections were made by the writer in various parts of the United States. The species about Chicago, especially scutellaris, were collected through the year from the same loealit.y with a view to getting the seasonal variation of the species and any variation from generation to generation. 405] COLORS OF TIGER BEETLES— SHELFORD 11 The color outogeuy woi'k was done on material dug in the larval stage at Gleiicoe, Illinois (C limbalis) ; at Gary, Indiana, (C tranque- iarka) ; at iliUer, Indiana (C lecontei) ; at Lyons, Illinois (C. pur- purea) ; at East Chicago, Indiana (C repanda) ; Chicago vacant lots (C. puncfuhifa) ; and Suman, Indiana (C stxguttata). These larvae ■were reared in a greenhouse in which the temperatiu'e was about 4 to 8 degrees C. higher than the out-door soil temperature. This accelerated the appearance only a little and did not show modification of color or pattern. The larvae were reared in sand, either in cylindrical lamp chimneys, setting in screen bottomed boxes or in screen bottomed boxes. "VSTien the majority of larvae and. pupated all were removed to small square watch glasses, lined with filter paper and moistened with 2% HjOo. These were piled up so as to cover each other and kept in a cool room, and watched closely to secure as many as practicable at the time of emergence. The elytral material was nearlj- all killed in a picro- sulfurie acid killing fluid and cleared and mounted in balsam, but some was preserved in glycerine jelly direct with good results. They were preserved at different intervals after emergence. The material for experiments was collected from the same places as that for ontogeny study and was subjected to high and low tem- peratures in an apparatus to be described later. The writer is indebted to Dr. C. B. Davenport, the late Professor C. 0. Whitman, Professor C. M. Child, and Professor W. L. Tower for suggestions during the first four years of the work (1903-1907). He is further indebted to the University of Chicago for funds amount- ing to .$400 and to Professor and Mrs. F. R. Lillie for funds amounting to $200 to cover expenses connected with the study and collection of the group. The Graduate School and Department of Zoolog.v of the Univesity of Illinois provided for later stages of the work. Acknowledgments are due Jliss Annette Covington for making the water color drawings of the ontogenetic stages and the other changes in color during the life history. Mr. K. Toda made the drawings of the geographic races shown in color and also the stippled drawings of ontogeny. The writer is especially indebted to Professor A. A. Jliclielsou of. the University of Chicago for making physical determinations of the colors of Cieindela. The courtesies connected with the study of collections were numerous. The following were especially kind in facilitating the study of collections in their charge : Dr. Samuel Hen- shaw, Museum of Comparative Zoology ; Dr. Sehwarz, U. S. National Muse'am ; Dr. Henry Skinner, Philadelphia Academy ; 'Mr. William Beutenmuller. American iluseum; Mr. Gilbert I. Arrow and Mr. C. J. Gahan, the British Museum; Professor Poulton and the late Mr. 12 ILLINOIS BIOLOGICAL MOXOGRAPHS [406 R. Shelford, Hope Collections, Oxford ; Pi'ofessor David Shai'pe, Cam- bridge ; Professor Kolbe, Berlin ; Mr. P. Lesne, Paris. The following permitted me to study their private collections with a considerable loss of time and attention: Dr. Walther Horn, Berlin; Mr. Basil G. Neviusou, 3 Tedworth Square, Lojidon : Mr. Gestro, Genoa. A considerable number of men whose names appear below gave me data on the distribution of the species in their collections, loaned or presented specimens, or did similar service through making exten- sive collections which were exchanged. For this I am debtor to Messrs. C. F. Adams, C. N. Ainslie, Geo. G. Ainslie, E. M. Anderson, A. W. Andrews, Germain Beaulieu, Biederman, Wm. Beutenmuller, Albert L. Barrows, C. S. Brimley, T. C. Brues, T. D. A. Cockerell, I. W. Cockle, Norman Criddle, F. F. Crevcoeur, C. C. Deam, G. M. Dodge, Edwin H. Edwards, J. D. Evans, S. A. Forbes, E. P. Felt, E. D. Har- ris, R. V. Harvey, G. W. Herrick, H. R. Hill, J. S. Hiue, A. D. Hop- kins, W. Horn, James Hunsen, S. A. Johnson, James Jolinson, W. Knans, Chas. W. Leng, H. P. Lodiug, D. E. Lantz, W. ^Macintosh, G. P. Mackenzie, L. E. Marmont, A. L. Melander, F. W. Nuuenmacher, W. E. Rumsey, L. E. Rieksecker, A. G. Ruthven, Franklin Sherman, H. F. Snow, E. G. Smyth, Tom Spalding, Chas. Stevenson, T. B. Sy- mons, H. B. Walden, H. F. Wickliam. T. N. Willing, R. S. Woglum, R. H. Wolcott, E. 0. Wooton, E. C. VauDyke, E. P. Venables, S. S. Visher. 407] COLORS OF TIGER BEETLES— SHELFORD ANALYSIS OF COLOR PATTERNS COLOR PATTERNS AND ELYTRAL STRUCTURES In the Cicindelidae usually onlj' the elytra have color patterus. These are merely sack like outgrowths supplied vAih. nerves, trachea, and blood spaces. The cutieular covering is in two layers;, the outer portion is a hard and relatively homogeneous layer known as the primary cuticitla and on the upper side is usually characterized by the presence of saucer-shaped depressions, somewhat hexagonal in form, fitting to- gether with common rims. These rims usually correspond to the posi- tions of the points of contact of the hypodermal cells and accordingly each cup corresponds to a cell (Packard, 1900 text). Some forms in the family, e. g., the Tetraehas and the Amblychilas do not have these cups; the surface is smooth. In certain areas the primary cuticula is pigmented and in certain areas clear and transparent. This gives the color pattern. Some species are almost entirely pigmented ; some entirely without pigment. Beneath the primary cuticle is the secondary cuticula which is laid down in successive layers during the life of the individual and in the forms like Ambli)chila cylindriformis, and Phaeoxantha klugi is essentially uniform in character. It contains some spaces, prob- ably pore canals, which are empty of cell contents except for the layer in actual contact with the cells. A few of these pore canals can be detected in the secondary- cuticula of Tctracha Carolina. In Cieindela the secondary cuticula beneath the pigmented areas of the elytron is clear and transparent and entirely free from the "pore canals" and interlamellar spaces, while beneath the unpigmented areas it is full of the "pore canals" and large interlamellar spaces, and these having been left empty by the retreat of the cells from the successive layers; they give the effect of a white or straw color depending upon the color of the secondary cuticula itself. In these regions, beneath the unpig- mented primarv cuticula, it is about twice as thick as beneath the pigmented parts (Fig. 1, PI. I). The color pattern may accordingly be described in terms of pigment and lack of pigment, the so-called markings being without pigment. The two walls of the sac-like elytron are held together by chitinous pillars or columns which in the adult appear in cleared elytra. The different lavers of cuticula show here as rings around the original 14 ILLISOIS BIOLOGICAL MOXOGRAPHS [408 central spindle (Shelford 1915:243, Fig. 1). In the Ciciudelidae the ehitinous columns are not arranged in any very defiiiite manner but in some eases they retain their pigment within areas that are not othei-- wise pigmented. Hairs which in a primitive insect usually cover the wing entirely are present in nearly all tiger bettle elytra. In the ilautiehoras, observed representative of the Pogonostomidae, and one of the Megacephalidae, Mcgacephala {Tetracha) aequinoctialis, the elytra are more or less com- pletely and uniformly covered with small hairs. Under the microscope the hairs may be located on the pigmented area of the elytra by the light area which is produced by the thin cuticula at the base of each hair. Hairs appear on the whole to be less common in the unpigmented areas and when present usually are surrounded by a narrow rim of pigmented cuticula. Hairs occur in i^ractically all groups, thoiigli they have been lost from the majority except for a few at the base of the elytron and scattered along the tracheae ( Shelf ord, 1915:243, Figs. 1 to 3). These are present in Cicindela and are sho's\ai by small circles in figures 2 to 29, plate I to III. The elj'tra of many species are marked with jjits. Close examin- ation under the microscope with both transmitted and reflected light shows that, in the majority of cases, the pits are over the center of the ehitinous columns and bear no relation to rudimentary hairs as Dr. W. Horn has suggested. I have seen uo pits that would appear to represent rudimentary liairs though they may occur. There are sometimes thickenings running lengthwise of the elytron as iu Domiea (Shelf ord, 1915: Figs. 35 and 36). While these thicken- ings run parallel with the trachea, they are usually between ratlier than coincident with them, except in Caledonica (Fig. 25). There are, how- ever, some thickenings on the under side of the elytra of most species which correspond in a general way to veins (particularly in Mantichoi'a ) . The outer and inner margins of the elytra are always thickened and resemble veins, almost invariabl.y containing tracheae. The subcosta usually follows tlie costa very closely at the base of the elytron but just behind the middle it turns inward away from the margin in a vein like thickening. The radius is in a distinct thickening of the elytron which proceeds from the base for a short distance. This is very constantly present. Aside from this nothing comparable to veins is present but the rows of ehitinous columns are often so arranged so as to give distinct and direct spaces running the length of wing. These are occupied by the principal tracheae. In some cases the spaces appear very clearly on the under side of the elytron and in JMautichora there are distinct ridges over them which have every appearance of veins. 409] COLORS OF TIGER BEETLES—SHELFORD 15 The elytral tracheation of the Cieindela has been observed by the writei" in about one hundred species. The elytra of the newly emerged ^magoes of ten North American species have been studied in some detail. Nearly all the common North American species and about fifty exotic species have been studied in less detail by mounting dried elytra in hot Canada balsam containing little or none of the usual solvents. The main tracheal trunks and some of the branches remain clearly visible in such mounts for several hours. In terms of the system of classification proposed by Comstoek and Needham, the usual tracheae present (Figs. 18 and 21, PI. II) are the costa {Co) which branches near its distal end, and subcosta (S) which lies close to the costa on the outer edge of elytron ; the radius {R) and media (.1/) which lie in the medium portion of the elytron; the cubitus (C») which lies along the suture, and {A) the anal rudiment which lies next to the scutellum. The six tninks common in insects are represented in but two genera (Aniblychila and JIantichora), which have rudimentary wings and specialized elytra fastened together in the adult (Shelfoi'd, 1915). These trachea are demonstrated in the adult dried elytra without any difficulty. In Omus, which is closely related to Amblychila, the radius and media have disappeared except for rudiments. The cubitus is the principal trachea. With the exception of Omus and Amblychila it is the anal that has degenerated farthest. Collyris was never veiy satis- factory for study, but it appears that the cubitus is reduced and the anal wanting. In Platychila pallida (Shelford, 1911: Fig. 7) the com- monest type of tracheation of the family and probably among the most generalized, so far as the first four trachea, are concerned, is shown. The anal is much reduced. The munber of small branches and cross connections is large and too variable to be correlated with other specific characters or with color- pattern characters. Figures already published (Shelford, 1913: Figs. 10 to 19) illustrate this fact. The two elytra of an individual show a mark- ed difference. It is evident then that only the main trunks are at all constant. The costal branch at the center the posterior third of the elytron at the beginning of the curve is very characteristic of Cieindela but bears an important relation to color pattern only in some cases. Figures 2 to 33, plates I to III are selected to show the relation of unpigmented ai-eas to the main traclieal trunks. Figure 2 shows four cross bands which are cut across by the tracheae. Figures 3 and 4 show the same type of pattern but with the cross bands narrower, the middle one broken in the region of the trachea toward the right and with a sug- gestion of two or three stripes. Figure 5 shows a similar condition but 16 ILLIXOIS BIOLOGICAL MONOGRAPHS [410 with the sjjots in the upi^er right-hand third of the elytron missing. Figure 6 sliows a suggestion of seven cross bands as numbered. Figures 8 and 9 are similar but somewhat broken and with some tendency to forming longitudinal stripes between the tracheae. Figure 9, Plate I shows a longitudinal row of spots and figures 10 to 13, Plate II either rows of spots or continuous longitudinal bauds between the tracheae. Figures 14 to 21, Plate III show forms that have lost most of their pig- ment and have retained it only in the lines of the tracheae. Figure 18 shows a form that appears to have double longitudinal lines between the tracheae and has lost the unpigmented areas in the anterior part of the elytron. Figures 16, 20, and 21 show forms that ai'e highly specialized as to the patterns and have lost most of the pigment and the media trachea is almost gone. It will be noted that there are many interesting curves and branches that are related to the color pattern. Figures 23 to 25, plate III^sliow an oblique joining of the markings to form a vitta that is not related to the trachea and is rather rare, con- stituting an exception to the usual rule. As a result of this study of the figi;res it is seen that in the color patterns of the genus Cicindela exists a system of markings that is related to the tracheae, and also is arranged with reference to the cross bands of which there are five, two of which may be, divided as to make seven and that these are arranged as follows: Therei is a cross baud in the center of the longest measurement of the elytron. This location is shown to be the same in essentially all of the cases by actual measure- ment. There is one at the tip and one at the base, with one or two arranged respectively between each of the latter and the middle one. These intermediate bands are most commonly represented as one but are some times divided, but in any case its center, or the center of the intervening pigmented area is half-way between the two adjacent, un- pigmented more permanent cross bands. It is also evident that there is a possibility of fusion of joining of light areas, so that these lines of fusion are in the spaces between the tracheae and in the region of the cro.ss bands. The areas near the hairs described in a preceding section are the very last to lose their pigment in the forms that become almost entirely without pigment. It is to be noted that in the forms that have the longitudinal stripes and cro.ss bands broken up, the media is almost entirel.y gone. It has been shown that these cross bands are the most constant wing markings in insects and are usuall.y represented as the i},ve first mentioned. I have gone over very large series of Coleoptera, (Tower, 1906), and find that tliis is true for tliis order, while cross bands in the Lepidoptera (Braun, et al. cited), Diptera. Ortlioptera, Tricoptera, Pleeoptera, Heraiptera, have been discussed by Von Linden, 411] COLORS OF TIGER BEETLES—SHELFORD 17 Eimer, et al. In the Lepidoptera (Mayer), however, the line of the veins is the one in which pigment is longest absent, but in the Diptera both living and fossil there is a uniformly denser pigmentation of the veins. Doctor Williston teUs me that it is true of the fossil forms, and Doctor C. F. Adams found in the development of the color pattern of some common ilies that pigment first appeared along the cross veins and spread from these. In the Hymenoptera the veins are often pigmented and the same is true of, the Mecoptera, Plecoptera, some Ilomoptera, etc. Pigment is usually found in muscle attachments and wherever rigidity is necessary; this has been reported by Tower (1906) in Coleoptera, and in Polistes by Entemau (1905). Since the veins are supporting structures, one Moidd expect that they would usually be pigmented. The great development of the secondary cuticula in the Coleoptera might since the el.ytra are no longer used as wings, .show modification characterized by the loss of this character in some eases. I find no observations on the secondary cuticula of the wings of Lepi- doptera. In the Ctenostomidae are found bands in some of all of these positions noted in Cicindela (Figs. 26 and 27, PL III, also 376, PL XYI.) In the Collyridae it appears that the band at the base of the elytron {!), one in the middle (4), and the one at the tip (7), are quite common and well developed (one or all). Collyris celebensis Chd. (Fig. 28) and arnoldi McL., horsfieldi McL., fasciata Chd. et al. have such bands. In Theratidae are found markings which conform to the cross bands of Cicindela, (Figs. 332 to 337, PL XVI), but the areas represented in the two ends of the elvtron niav be much extended (Figs. 236 and 237, PI. XIII). Turning to the other form of the Cicindelidae proper, one finds that in the Euryodini and the Odontochilini markings occur in the same relations to structures as those already described. Among the Euryodini, in Caledonica occur some of these cross bands indicated, and in addition a very interesting thickening of the elytron in the lines of tlie tracheae (Shelford 1915: Fig. 25). These may or may not correspond to the thickenings that are associated with the veins of other insects, for iu the Dromicini. (Cicindelidae proper) we find thickenings that lie be- tween the veins and may be regularly arranged (1. c. Figs. 35 and 36). It will be noted that there are spots in places corresponding to those already mentioned. for example, crossbands in figure 29 repre- senting the Odontochilini, and longitudinal stripes in figure 30 repre- senting the Dromicini. In the Megacephalidae the color patterns are in some cases like that 18 ILLINOIS BIOLOGICAL MONOGRAPHS [412 shown ill figures 370-372, plate XVI, which accord with those of Cicin- tlela. Ill Megacephala klugi we find a curious dark spot in the position of the cross veins between the subcosta and the ramus which corres- ponds to the condition that Tower (1906) has called attention to in the Coleoptera, Lepidoptera, etc., but it is not of frequent enough occur- rence to be significant. I know of no color patterns in the Palaeomau- tichoridae or the Neomantichoridae/ both the wings are rudimentary and in the latter the eyes are much 'reduced and they are in some eases light avoiding. In Platychila pallida we have only a very slight pigmentation any- where on the body ; the wings are reduced to a rudiment that is barely distinguishable and the elytron is pigmented only in a small area lying in its anterior two thirds and along its inner side. There is no develop- ment of spaces in the secondary cuticula sufficient to make the chitin opaque and yellow. In the Dytiscidae, Carabidae, and Haliplidae, the chitinous columns are arranged in definite rows and likewise in many cases the hairs and glands. The center of these chitinous columns, or better the primary cuticula over the chitinous columns, is last to lose its pigment; accord- ingly one may find a line of pigmented spots lying in rows, often two rows, between the tracheae, for example as is shown in the Bembidium versicolor Lee. (Pig. 35). The row of chitinous columns break across the white markings and in some of our common Haliplidae, for example, the chitinous columns are so arranged and the centers are associated with the openings of glands, the cells of which have caused the column to be cut half in two. To find what are the conditions of the tracheal structures in other Adephaga I made an examination of a number of forms in the Dytis- cidae, Carabidae, and Haliplidae, (Figs. 34 to 41, PI. IV). Omophron shows all six tracheae and three cross bands which do not appear to be related to the tracheae. Bembidium versicolor shows only five tracheae, but the unpigmented areas are in the lines of the tracheae and also between them. Nebria complanata (Fig. 37, Europe) shows the tra- cheae in the lines of pigmentation as well as a suggestion of the double banding shown in the Dytiscidae. The Dytiscid {Hydacticus stagnalis. Fig. 38) shows all of the six tracheae and a light line both between and directly above them. Figure 39 (Laccophilus rnaculosus) shows suggest- ions of cross bands and double stripes. Agabus temiolatus (Fig. 40) shows the tracheae within the lines of the unpigmented cuticula. Hydro- porus undulatus (Fig. 39) has the cross bands and the tracheae appar- ently between the spots. (Compare with figures 2 to 25, plates I to III). From the studies preceding, especially the last, it is observed that there is no constant relation between the tracheae and the distribution 413] COLORS OF TIGER BEETLES— SHELFORD 19 of the majiings or iinpigmented areas, of such a character as to suggest a direct pliysiologieal relation between the two. In the specimens in which the tracheae are unusually arranged there is no effect on the color pattern or variation on that suggests a direct relation between the two. Nor is there any connection between the oxygen supply from the tracheae and the pigment. And as the blood sinuses and tracheae are for the most part coincident, I see no reason for relating the blood supply to these characters. The folding of the elytron in the pupa is apparently not related to the cross bands. It accordingly appears that the relation of pigment formation in the elytron to structure is not directly causal, at the present stage in the evolution of the groups but is one belonging to the general structural organization, hereditary in character. THE COLOR PATTERN PLAN The pattern of the Cicindelae is analyzable into the areas or tend- encies sho^^Ti in figures 42 to 49, plate V. Figure 42 shows the full number of dark and light longitudinal stripes. The light stripes are labeled a,A,B.C ; a is not usually distinct. Very often it is absent as in figures 3, 4, 6, and 11, but sometimes appears to be present without A as in figures 5 and 13, plate I and II. It is often present and partially separated from A in an Australian species (Pigs. 50 and 51, PI. VI) only. This Australian species is the basis for figure 42. More often it is joined with A (Fig. 43), and not recognized separately (Fig. 52). Figure 44 indicates a tendency to double lines between the tracheae suggested by an African species (Fig. 53, also 57 and 7 and 8). Figures 54 and 56 show the longitudinal stripes partially represented. Figue 45 shows the full niunber of cross bands rarely complete numbered 1 to 7 ; but perhaps best represented in figures 57 and 59 to 63 where they occur broken two spots. Bands 5 and 6 occur nearly complete oftener than 2 and 3 (Figs. 57 and 75). Figure 46 shows the type in which 2 and 3, and 5 and 6 are fused. This is almost a duplicate of the pattern of an .\frieau species, figure 58, but also well represented bj- figures 73 and 74. Figure 47 shows a common type, cross bauds 5 and 6 being separate but the more anterior ones being reduced at the anal side of the elytron. Figure 48 shows all the possible spots which resulted from the superposition of the longitudinal stripes and cross bauds. There are 19 of these, of which 11 occur in an Indian species (Fig. 62). Figure 49 shows the spots or elements from which the characteristic patterns of the group are made up. This pattern should be compared with figures 31 to 33, plate III. which show that individual variations follow the rule of the entire group. The usual pattern of C. tranque- 20 ILLIXOIS BIOLOGICAL MONOGRAPHS [414 Varica Herbst is seen to be made up of Al, A2, B3 {humeral luniilc) ; A4, B4, B5, Co {middle hand) ; and A6 and 7 {apical lunule). In figure 31 may be noted the forward hook-like extension on the so-called humeral lunule which represents B2, the union between B5 and C6, etc. Figures 61 to 72, plate VI indicate some of the commoner combin- ations of spots. Figure 66 shows a union between the humeral lunule and A3 ; figure 70 a combination of 7, with B6 and B5 which are connect- ed with the central cross band ; figure 71 a cross connection in baud 2-3 between stripes A and B ; figure 72 a cross connection in band 4 between A and B ; figures 76 to 77, plate VI show the reduction of cross bands to large spots. Thus the conclusion that the patterns are derivable from combination, loss, and extension of a number of inter-tracheal spots fall- ing in cross rows seems justified. There are various types of combina- tion and extension which are not common when the group is considered as a whole, but which represent tendencies in certain isolated groujjs of species and which must be illustrated (Figs. 78 to 98, PI. VII) here because they othewise appear to be obstacles to the plan. One of these tendencies is one toward oblique combinations indicated in' figure 78 (a diagram). One type indicated by the wide stippled band is shown in figure 79, a South American species. A similar combination occurs as a variation in an Indian species (Fig. 80). A more gentle sweeping combination is shoA\Ti by the narrow white line in figure 78 and occurs as the regular pattern of an African species (Fig. 81) ; shorter curves occur in another African species (Fig. 82). Other oblique combinations are shown in figiires 83 and 84. The type of obliqueness shown by sev- eral African species (Figs. 85 to 87) is an oblique shifting of the entire pattern ; it appears to be turned parallel to the end of the elytron. This appears to be a significant tendency and will be discussed again in con- nection with the discussion of experimental results. Figures 88 to 91 show a tendency toward obliqueness of markings reversed as compared with that just described and characteristic of the princeps-cylanensis gi-oup of India and Africa. It may be said to char- acterize the patterns of a group standing apart from the other represen- tative of the genus. Figures 92 to 95, plate VII show unusual sinuate extensions of the markings. In figures 92 and 93, Indian and Australian species, a mark- ing resembling the usual "middle band" arises in the area A2.3 with a form similar to that found in figiires 94 and 95. Figure 96 shows bands 5 and 6 separate toward the outer margin of the elytron and united toward the inner. Figure 97 shows unusual extensions of the markings giving two light bands between the tracheae (compare with Fig. 19, PI. II) ; 98 shows unusual direction of extension. From the preceding discussion and diagrams I concluded that even 415] COLORS OF TIGER BEETLES—SHELFORD 21 the most complicated patterns are redncable to the usual plan or are made up of unusual combinations of spots occurring in other groups of species. Certain laws regarding direction of shifting of markings seem to prevail. These will be noted again in another part of the paper, (page 58). COLOR PATTERN AXD PIGMENT DEVELOPMENT As an example of the usual type of pigment development in Cicin- dela let us follow the events in C. tranqueharica (PI. VIII). In the yoiingest pupae there is essentially no pigment present except sufficient in the eyes to give a slight brown color. This gradually becomes darker until tlie end of about ten days when the eyes are a dull brown and the process is apparently complete. At the end of 12 days the tarsal claws, the tip of the mandibles, and the tips of the mandibular teeth have received their full quota of pigment ; the pigment proceeds from the tips proximally and by the 13th and 14th day pigmentation is com- plete. On about the 13tli day the distal portion of the tibia of all of the legs show pigmentation on the outer side and this proceeds to the more pi'oximal portions most rapidly on the outside of the leg. The most distal parts of the tibia are pigmented about 2 or 3 days later. Coin- cident with the development in the tibia is the development in the trochanters where it begins at the outer margin. A slight darkening takes place in the mid-portion of the developing hind wing which is so folded as to make the tip of the pupal wing show dark. At this time, viz., at the end of from 14 to 16 days, the insect emerges. Often at or before the time of emerging the first color centres of the dorsal side of the abdomen have appeared on the last abdominal segment and more rarely also the corresponding centers of the next to the last segment are also present (Fig. 105, PI. VIII). Usually the animal en:erges with the tibia, tarsal claws, part of the trochanters, eyes, mid-portion of the hind wing, and tips of the mandibles pigmented (Figs. 101 and 105a). The later histon- exclusive of the elytra is as follows: Tlie pig- mentation begins first on the distal joint of the antennae and the max- illary palps (Fig. 101), and on the teeth of the maxillae. After about 8 hours the tip of the inner palp and the ligular portion of the labium shows pigment (Fig. 102) ; next after about 12 liours the distal segment of the labial palp and the outer wings of the labium darken (Fig. 103). The gula begins to show pigment about as soon as the ligula, and the pigmentation of this part is complete at the end of 12 to 15 hours. At tlie time (after 12 to 15 hours. Figs. 103 and 107) the general pigmenta- tioji begins to be most rapid, pigmentation begins to show strikingly at the proximal portion of the appendages just noted and proceeds to 22 ILUXOIS BIOLOGICAL MONOGRAPHS [416 meet the distal pigmentation (Zeleny, 1907). The extent to which it goes differs in different species and gives a faint pattern to the parts in some species. The pigmentation which begins distally, usually pro- ceeds only through the extent of the more distal segments (Fig. 109 a to e). By the end of 24 to 36 hours (Fig. 104) the pigmentation is nearly complete by development over the general areas of both body and appendages. Thus in the antenna at the end of 8 to 10 hours rings appear toward the distal end of the tliree proximal segments, darken and spread toward the proximal ends of the segments rapidly (compare Pigs. 109 c, d, and e). The pattern shown in the antennae, legs, mandibles, palps, etc., persist in some species (see page 24). At 3 to 6 hours after emergence (Fig. 105a) the suture between the clipeus and head becomes pigmented. By the end of 8 hours after emergence there are two oblique color centers between the centers of the eyes; these correspond in position to the oblique depressions that occur in the genus Tetracha. Beside these there is a center close to the posterior side of each eye, one just behind and inside of this, and one in the middle of the frons (Fig. 106). Pigmentation then has proceeded backward on the clipeus, and backward from the suture of the cUpeus on the head (Fig. 106). At the end of 12 to 15 hours, the pigment of the clypeus and anterior part of the frons and centers just described has increased and extended backward, giving a pattern as shown in figure 107. This process continues with general suffusion over the head with the pattern still in evidence at the end of 24 to 36 hours (Fig. 108). After 8 to 10 hours after emergence (Fig. 106) the posterior border of the thorax shows two centers in the depression at the posterior side. Little change takes place on the under side from emergence. By the end of 12 to 15 hours (Fig. 107) the thorax has presented some new centers, a longitudinal stripe occurs near each margin, and there is a narrower one between each of these and the center, and the anterior depression is darker than usual. The end of 24 to 36 hours (Fig. 108) shows the obliteration of the centers mentioned above by tlie pigmenta- tion of the inter-spaces. On the ventral side of the abdomen and thorax pigment begins on the outer side of the more posterior segments first and centers appear from behind forward. During the first few hours the pigmentation does not begin on the remainder of the abdomen. The next center to appear is the one in the center of each segment near its anterior side, which appears between the 6th and 10th hour. Just a little later a line appears across the posterior side of the segment and there is an exten- sion of the center one at each side and the coming in of the a loop-like addition outside of the first center. This svstem of markings is best 417] COLORS OF TIGER BEETLES— SHELFORD 23 understood b.y a eomparisou with the larval segments (Fig. 99 a, h, and aa). If a and b joined to give the first marking that appears, «a stand- ing out clearly and all of the rest joined laterally, one would have the condition found in the development of the adult color. No change takes place in the thorax except the development of a center of. tbo on the middle line of the meta-sterniim which probably repi-esents the attachment of the large hind wing muscles (Fig. 102), until the coloration of the abdomen is has been completed in the veuti'al side of the third, fourth, and fifth, and last abdominal segments; this j having proceeded from behind forward. At the end, 12 to 15 hours 0' i (Fig. 103), it will be noted that the hind coxae, the ante-eoxal pieces, ] the epistema of the metathorax and the coxae of the other segments have received a quantity of pigment and a new; center has developed behind each metathoracic leg on the metathoracic sternum. The next stage represented (24 to 36 hours, Fig. 104) shows a general diffuse pigment on the entire ventral surface except the outer sides of the metathoracic coxae which long remain unpigmented. The ante-coxal piece is nearest complete. The great possibilities of being deceived as to position of the color centers is shown by the fact that the abdominal centers and center behind the legs on the metathoracic segments is lost entirely in the last stage of the development. Conditions on the dorsal side are very simple and centers appear just as in the larvae (Fig. 100) two in number on each segment, begin on the last segment, and move forward fusing in the middle line, and in course of about 10 hours after emergence (Figs. 106 and 107) the color of the dorsal side of the abdomen is practically complete. In regard to the color centers of the ventral side of the abdomen it may be said that they are the same in number and arrangement as found by Tower on the ventral side of the abdomen of the potato beetle larvae. The abdominal centers are serially homologous. The pattern , of the dorsal side of the abdomen of the larvae if Leptinotarsa is similar ^/ to that of the ontogenetic ventral of the adult C'icindela. The upper side of the abdomen in the potato beetle larvae is divided with respect to these structures because growth, bulging, and wrinlding due to the extension divide the dorsal side into two parts, and have resulted in the separation of the centers into two rows or bands (see Tower. 1906: PI. 18). In the larval Cicindelidae, however, it is the ventral side that is extended in the process of the development and which may be wrinkled and the centers are separated just as in the case of the dorsal side of the abdomen in the larvae of Leptinotarsa. There is never any tendency for the dorsal side of the cicindelid abdomen to wrinkle ; in fact it is reduced as compared with ventral. On the ventral side of the adidt Leptinotarsa abdomen six centers appear but these are not divided as to the middle hi 24 ILLIXOIS BIOLOGICAL MONOGRAPHS [418 of the segment. Tower's basipleural is no doubt represented by the three spots that are near the spiracle (Fig. 99, PL VII). In the prothorax of the tiger beetles it is to be noted that the onto- genetic coloration is parallel to that in some of the Leptinotarsae ; the two pairs of parallel lines which occur appear to eorresiDond to markings that are on the prothorax of the C. tranqucharica (Fig. 107). The two oblique centers of the frons or epicranium in the Cicindelidae are rep- resented in the Leptinotarsae also the two markings by the eyes. I have noted that in the aiitennae centers arise in the form of rings around the distal ends of the 2d, 3d, and 4th segments (Fig. 109 c, d, e.). Conditions in figure 109d and c show patterns in the tlevelopment of these which are the exact duplicate of the patterns in the antennae of the G. strachani (Africa) which has also a primitive elytral pattern.' C. theratoides (New Guinea), many of the Megacephalidae and some Collyridae. The development of the pigment in the legs up to tlie time of cmergencq is described above ; after emergence the development pro- ceeds from proximal to distal in the tibia and in the same succession in the tarsal segments. Previous to emergence the humerus is somewhat compressed and wrinkled, being only about two-thirds as long as after the expansion which follows emergence. At the end of about 8 hours one finds the feruAr beginning to show a general suffuse pigment which appears to arise snnultaneously over the entire surface. After this the later liistory in tlie legs im simply a general intensification of the pig- mentation. In all of the species of Cieindela studied the phenomena of pigment development are the same so far as has been noted above with the exception of the ptinctulata and lepida in which the first centers appear in the middle of the ventral side in the third and fourth abdominal segments. This is tlie case in T. Carolina in which the centers are like those in the larvae. The adult abdomen in this species is not pigmented toward the posterior end of the ventral side while the upper side never receives any pigment at all and the usual larval color center are, as has been stated, very much reduced in this species. Likewise the centers of the head and prothorax are little developed and the obliqiie ones near the center of the frons are very faint. The two which appear first in the posterior depression of tlie prothorax are quite distinct and very suggestive of the condition in the MegcicrpJuda (Phacoxantha) khigi. The legs are, however, not pigmented at all and it appears that the cuticula in these cases is of the type with interlamellar spaces which is a means of giving strength to the less rigid parts of the body. An examination of stained whole mounts of the appendages shows that as a rule the distal portions are first clearly differentiated and 419] COLORS OF TIGER BEETLES—SHELF ORD 25 iirst to take on the form that the part is to have in the adult. The tip of the mandible is the first to sliow the distinct pointed form toward the head, tooth after tooth being differentiated from the somewhat larger mass of tissue which makes iip the mandibular outgrowth. The same is true of the other mouth appendages, labrum etc., the.y become more hairj- in form and stand well separated from the old pupal skin. In the ease of the leg the tarsal claws are the first differentiated and this pro- cess appears to move in a general way toward the body, segment by segment, the femir being last to be differentiated and last to receive the (/^ / pigment. The position of the pigmentation in the tibia corresponds to the point of attachment of the flexor of the tarsi. This is early developed and thus the tarsi are the first to become movable ; at this time the flexor and extensor of the tibia are not well developed, their muscle striations appearing indistinct, but become much more distinct and definite in form a little later and the tibia becomes movable about the time of this development of its pigmeut. A similar development occurs V on the proximal portions of the trochanters which are the attachment of . the muscles. The wrinkled condition of the femosa helps to give it t^-V/ rigidity and. the legs are well enough developed to allow of sufficient movement to release the animal from the pupal skin. The legs are at first somewhat extended and subject to a considerable amount of move- ment, and while the body is flexed and extended and the pupal skin ruptured in the midline of the thorax the mandibles are then worked as well as the other mouth parts and the head removed by repeatedly throwing it backward. The animal gradually wriggles out of its skin and the wings and elytra soon expand ; the wings expand to the full length inside of about 20 minutes after the animal emerges, and remain thus for several hours. If for any reason the expansion of the wings or elytra is interferred with, they always remain in the exact condition in which they were placed by the adverse conditions, and if the wings are not folded in the normal fashion at the proper time, they will always remain completely extended. Their early pigmentation, if it is associated with hardening, is probably an advantage to this process of withdrawal or folding. It seems altogether probable that the peculiar manner of develop- ment of the pigment is associated with the development of the structures which are necessarv' to ectosis and that they accordingh- represent de- velopmental adaptations. In the ease of the tiger beetles which do not have the appendages pigmented in the adult the cuticvila must harden without being pigmented. The animals emerge M'ith the elytra entirely unpigmented and dur- ing the first 4 to 8 hours little change is easily noted. One can hardly record the beginnings of the pigmentation as this is verj' faint, and the 26 ILLIXOIS BIOLOGICAL MONOGRAPHS [420 wings beneath, which come between the elytron and a part of the pig- mented abdomen, give no opportunitj' for accurate observations with the elytra in position except as one slips pieces of paper under them, which may injure the elytra so as to give abnormal development. The elytra must be removed and mounted in glycerine jelly, or cleared in balsam. It is necessary to hold the slide in position over the surface of a good glass plate that has been painted white on the lower surface and uot magnify them or if so, only about two diameters with a reading glass. The fresh, unmounted elytra may be placed in formalin in a watch glass painted a neutral gray or yellowish tone which is the same color as that presented by the elytron before pigmentation when viewed in transmitted light. By this method and with individuals killed at different stages, and with the use of a Zeiss binocular microscope, I have been able to follow the course of pigmentation of the elytra. The elytra have been examined in cross section ; there are no thickenings in the primary cutieula in which all the pigment is located, except the small thickenings that have been described as occurring in the area immediately in front of hairs, and these have been carefully considered and their relative number as effecting the color effect practically elim- inated. The cutieula is somewhat thinner at the tip of the elytron. The actual hairs present are surrounded by an area that is fully pigmented, but this also has been taken into account. Elytra of C. repanda show beginnings of pigmentation which often are strongest near the costal border at the end of 4 to 5 hours (Fig. 111). The chief of the areas showing lack of pigment are in the lines A and B and are particularly prominent near the base. Later (Figs. 112 and 113) these lines are broken into spots which correspond to spots found in certain Eurasian and African species (Figs. 147 to 187, PI. XII, and 241 to 280, Plate XIV). The series of stages that I have had has been small and not siaited to the detailed comparison as some of the following sjjecies are, but shows the same thing. The color development in C. lecontei Hald. begins very faintly ajiparently at about the posterior end of the anterior third of the elytron, at first the permanent markings are difficult to distinguish, but a little later they become distinct patches. Two ontogenetic markings between the base of the elytron and the general arrangement of pigment at the end of 4 to 5 hours (Fig. 114) correspond verj' closely to conditions found in repanda. Longitudinal, heavily pigmented stripes that stand out in some individuals, lie in the lines of the tracheae and hairs, and become more pronounced as the development continues. Figure 115, 12 hours after emergence, shows none of the spots characteristic of the others shown but has indications of a cross band which never occurs in lecontei but which is present in rugifrons and modesta of the Atlantic 421] COLORS OF TIGER BEETLES—SHELFORD 27 coast. Figui'es 116, 117, 118 sliow a number of spots arranged in longi- tudinal rows. A comparison of these witli figures 156 to 177 and 244 to 261 will make clear a close correspondence between the spots appearing and those in adults of Eurasian and African species. Pigment fails to develop when tlie elytron is wet (Gortuer, 1911). This happened in practically all wet elytra of this species and very few of those in other species. Development of pigment in the hind wings begins a little back from the anterior end, and in this case, about the time of emergence (Fig. 119), in the region in which the folding occurs, and shows while the wing is in the pupal skin thus causing the tip of the pupal wing to look black. Pigment passes out along the veins in both directions and vein after vein is pigmented toward the anal border. This process requires several daj's for completion. Figures 119 to 122 show the wing from the time of emei'genee to the end of about 24 to 36 hours and the adult. Development in C. purpurea Oliv. var. limbalis Klg. (PI. X) per- haps shows more definite spots than any of the others. The first evidences of the pigmentation in ontogeny is in the small circles around the hairs on the elytron; this takes place about 3 hours (Fig. 123) after emerg- ence. At the end of 8 hours the pigment usiially begins to come in gen- erally, first, in the lines of the tracheae. As in the case of the lecontei the first trace is at the posterior end of the anterior third of the ely- tron. The principal early developmental markings show as large light areas (Figs. 124 and 125, aftei^ 8 to 10 hours) which seem divided again later (Figs. 126 to 129, PI. X) and correspond to the spots found in old world species, figures of which have already been cited. Heavier pigmentation often persists in tlie line of the tracheae even in the adult (Fig. 130). In C. tranqucbarica (PI. X) the pigment begins first a little ])ehind the anterior end, as in the other species, and comes in the lines of the tracheae with all of the bands represented and the spots growing fmaller and the longitudinal stripes less and less prominent as time goes on. In all tlie elytra, liowever, the same markings appear as in the other species (Figs. 131 to 134), and spots occurring in other species are consistent in occurrence. C. punctulata (PI. XI) begins pigmentation about 4 to 6 hours after emergence and the pigment appears to pass from the anterior to the posterior end of the elytron. Certain lighter areas appear especially at the base of the elytron and between the tracheae, figures 135 to 137. These represent cross bands and otiier bands occur further back appear- ing in some cases but all are comparatively indistinct. There is, how- ever, a different phenomenon such as occurs in some of the Dytiscidae, 28 ILLIXOIS BIOLOGICAL MONOGRAPHS [422 e. g., LacchopMlus maculosus (Fig. 39, PI. IV), a coucentration of the pigment around the markings. Even where the markings are absent or ahnost so the denser pigmentation is present. This seems to have obliterated ontogenetic markings as they are shown less plainly in these species than any other studied. Such spots occur in some species of Cicindela as for example campestris, aulica polysita, latrcillei (Fig. 257, PI. XIV) and ismenia (Fig. 366, PI. XVI), which have a more densely pigmented spot in the region of the sutural spots of other species, i.e., in the position of C2.3 (Fig. 48, PL V). In cleared elytra of campestris a dark area appears at this point. Elytra of C. limbalid (Fig. 127, PI. X) shows this. In some cases dark .spots appear at this point in surface view; in others metallic spots. When the dark color occurs, the conditions described in page 51 are reversed — the surface film is absent. The distribution of the ehitinous columns above which areas are first pigmented makes the study very difficult. The hairs on the elytron which lie in the lines of the tracheae show pigment around their bases by the end of 3 or 4 hours if not earlier. The elyti-on reaches the adult color so far as pattern is concerned at the end of about 15 hours, but pigment continues to be deposited for several days. Only one stage of C. sexguttata (Fig. 138) studied shows the spots in the area between the tracheae faintly. The[ pigment is piled up about the markings only to a slight degree. C. punctulata and sexgut- tata belong to one of the Mexican groups and differ from the other species studied. One specimen of Tetracka Carolina (Fig. 139, PI. XI) was studied; in this the pigment began to develop at the end of about 9 hours and to manifest itself at the outer side of the elytron where it bends under, and appears to move toward all parts of the el.ytrou from there. A somewhat lighter streak was left, however, between the costa and the subcosta tracheae; this corresponds to stripe a, figure 139. The pig- ment moves toward the inner angle but shows a lighter space at the base between the ramus and the media and also a longitudinal stripe between the media and the cubitus, which is broken at a point corres- ponding with the dark band B between 2.3 and 4. This same break occurs in the area between costa and subcosta. That portion of the tip of the elytron between the media and the suture is the last to be pig- mented. Figure 140, which represents the elj'tron at the end of 9 hovirs shows adult coloration. The darker dots represent the ehitinous columns over the center of which the primary pigmented cutieula is thicker than any where else. At the point where it has been stated that (^ ; the pigment began developing the citicula is somewhat thicker than elsewhere. 423) COLORS OF TIGER BEETLES— SHELFORD 29 In C. hirtkoUi^ (Figs. 141 to 145) the pigment appeal's to begiu almost uniformly over the elytron except for the weaker places repre- senting tlie ontogenetic markings. The lighter places are between the lines of the tracheae. There are cross bands at the base of the elytron, the middle one usually moi'e or less clearly connected with the distal end of the adult cross band 3. Usually there is a spot opposite the end of this band between the media and the cubitus and usually another set of dots stretches across the elytron between the band 2 and band 4. These bands of a secondarj' nature are not present in the later stages or if so not marked. The longitudinal lines become weaker as time goes on and the markings, except those that ai'e to be permanent, gradually disappear; those in the region of the base are last to go. In some cases lighter longitudinal lines are divided into spots. A late stage in C. 12 guttata shows the same longitudinal stripes and cross bands. Throughout the series longitudinal stripes seem to be most marked in the earlier stages but become partially divided later and are rarely or never continuous but nearly always broken into spots. This is shown in nearly all the figures presented and the conclusion which seems warranted is that the longitudinal stripes are a more defin- ite character than the cross band, though neither occurs alone. The fact of a combined cross and longitudinal system of unpigmented areas is the one which comes forcefully forward in the entire study though there are irregularities present. Further, one sees a close resemblance between the ontogenetic patterns and those of the African and Eurasian species on which the analysis of the pattern was based. One notes also the close correspondence between the spots sho^vn in the general plan presented in figure 48, plate V, and tliose occurring in the ontogeny of the patterns of common North American species. This would seem to establish the plan of the pattern as well as could be hoped. The entire set of evidence presented tends to show that the simplest type of pattern in the Cieindelas is a pattern of spots lying in lines be- tween tile chief longitudinal tracheal trunks and falling into cross bands of which there may be seven. In ontogeny these are subject to some variations but such a description fits the general relations found better than anything else that can be stated. Such a type of pattern, which is of the character that is commonly called primitive, is what might be expected among insects. The wings are usually characterized by longitudinal veins which are thickened and hardened and often pig- mented. These veins are connected transversly by cross veins which are much more diversified in the insect group than are the longitudinal ones and which are also much more subject to individual variation. Tracheae usually occupy tlie longitudinal veins but not always tlie cross veins, hence in the insects which have actual cross veins there is not a neces- 30 ILLISOIS BIOLOGICAL MONOGRAPHS [424 sary correlation between the veins and the tracheae. The greater hard- ening and more general pigmentation of the veins of many insects already mentioned (page 16) leads to a spotted type of wing, in many cases at least. Such a system oiit'ered in the elytra of the tiger beetles gives the basis for the spotted type of elytron which we tiiid frequently in the group. Veins no longer occur definitely longitudinally and the tracheae do not ordinarily bear any definite relation to cross areas. A large background of evidence is presented above for the selection of the spotted type of tiger beetle pattern, made up of spots falling into rows and forming sti'ipes and rowS forming cross bauds, as a general one from which other types are derivable iy the loss of spots, com- iination of spots, etc. Comparable analyses were presented by Eimer (1895) and Von Linden (1902), Vv'ho note cross bands as the basis of the patterns of various species of Lepidoptera. Tower (1906) reduced the general plan of markings in Leptinotarsa to cross bands and longi- tudinal sti-ipes. He recognized 4 or 5 uupigmeuted cross bands and 6 longitudinal unpigmented stripes which fall in the lines with the tracheae instead of between them as in in Cicindelidae. He shows the stripes divided into two in the area between the costal and subcostal tracheae (Tower, 1906:228, Figs. 5 to 8, PI. XXIV), which is com- parable to the condition suggested in the carabids and dytiscids shown in figures 35, 38, 39, and 40, plate IV. Tower adhered to a theory ofteu held by embryologists, namely that the base of the wing is oldest ; further, that pigment appears first in the base of the elytron and pro- ceeds to the distal portion in accord with the relative age. No conclusive evidence is brought forward to show that the base of the elytron is actually oldest, and an examination of Tower's figures (Tow- er, 1906: 156, Figs. 1 and 2, 7 and 8, PL 19) shows that the basal part of the elytron in some species is not first pigmented. Pigment begins in the costal border of the wing and at the level of the second dark cross band which he calls the "proximal" and which is very common in his group. This is comparable to the early stages in Cicindela (Fig. 111). The view that pigment comes in first in cuticula over the oldest tissues from the embryonic standpoint seems not to hold good in Cic- indela, for on this basis certain abdominal sclerites would be embry- onicly older than others (Figs. 102 to 104, PI. VIII), the last abdominal segment older than the first, and the femur younger than the tibia as well as other peculiarities shown in figures 99 to 103. The law cannot be said to hold good at all in the group under consideration, but rather as has been noted on page 24, there is an order of post embryonic development of adult organs, which coincides with pigmentation. One of the most recent color-pattern analyses (Braun, 1914) shows the pattern of Lithocoletes (microlepidoptera) to be made up of a mod- 425] COLORS OF TIGER BEETLES— SHELF ORD 31 ication of seven transverse dark bands with six transverse light bauds between them (page 161). The figure of the hypothetical pattern is in general terras almost identical with that shown for Ciciudela and inde- pendently conceived on plate I, figure 4. In this second and third light band are represented by a single wide one and the fifth and sixth are separate as two narrower bands. If the general plan of longitudinal and cross bands in insect patterns is to be accepted we must also con- clude from the evidence presented that the relations to trachea, may be - reversed, i. e., the pigmented areas may lie immediately above the tracheae or between them. In the Lepidoptera pigment appears last in the veins (JIayer, 1896). The areas between the trachea may be subdivided into two longi- -W tudinal bands. The pigmented and'unpigmented bauds may also be reversed iu position as would appear to be the case when we compare the usual cieiudelid patterns with those studied by Tower. There is no reason why this should not be the ease as when markings are lost; the pigmentation which results is often heavier than elsewhere (Figs. 135 and 136). However when one compares the cicindelid ontogeny with the exist- ing patterns of other orders one finds that they show a series of light spots sncli as might easily correspond to the so-eaUed cells or areas divided by longitudinal and cross veins in a primitive insect such as a may-fly. The may-flies, stone flies and many diptera show such an arrangement in some parts of the wing. At least it may be safely con- eluded that a pattern of faint spots is the primitive type in Cieindelidae if one accepts any of the current criteria for primitive forms. I start with this type of pattern as "primitive" with a conscious- ness of the fact that it would be possible to proceed in entirely different directions and from entirely different starting points and make out cases of modification in definite direction fully 'as plausible as the ones here presented, provided only the preceding strong evidence is not ac- cepted. On this account it may be well to give the reasons for presenting this matter of modification at all. First, it is presented to further establish the contentions already made as to the character of the pattern plan presented ; secondly, to show that all even the most specialized types of patterns coxdd have been derived from the generalized types described above: thirdly, to show that there are certain laws of modifi- cation which nuist have been very general in tiie group and which have operated again and again in the production of the characteristic types of patterns. Figures 149&, 156a, and 165, plate XII, show some of the pat- terns in which five nearly complete cross bands occur; 179 shows a very 32 ILLINOIS BIOLOGICAL MONOGRAPHS [426 simple band, 1 with a complete, 4 etc.; 185 shows a wide cross band representing 3 and 4. Figures 149&, 149a and 156 represent the patterns of an African species showing tliat variations are in the direction of greater obliquity of the cross mai'kings, 149 approaching very closely to 148, wliich is a different species and usually oblique. A third species is sti'ikingly oblique but still possessing the usual cross bars of the group of species. Thus in this small group the usual typical pattern as shown by the general observations preceding is decidedly distorted by in a defin- ite dii'ectiou. In figure 165 is shown a type of pattern in which the cross bands are nearly vertical to the inner line of the elytron ; all the spots present fall in to such bands as they do in the ontogeny series (Figs. 112, 113, 116 and 117, PI. IX; 128 and 133, PI. X; 143 to 146, PI. XI). In all the other figures on the upper half of the page the two spots near the elytral suture are not in line with the cross bands. Evidence of this will also be found in the ontogeny series but is less marked than the tendency toward transverse bands. In 165 and 165a and in 156 and 156a, plate XII, the components of crooked niiddle' band are clearly brought out in course of variations in which two bands may or may not be joined in the stripe between the media and radius. This tendency should be noted as the most characteristic of the genus Cicindela, as there is scarcely a group of species as arranged on the basis of pilosity by W. Horn in which some one does not show this type of joining. The breaking of the cross bands by pigment in the line of the media is also very characteristic, but the tendency for the spots to lie out of the lines with the cross bands as interpreted, is taken as evidence of one of the general tendencies to be discussed later. The relations of the characteristic patterns to the general plan is .thus made evident. An- other general tendency also manifested is the tendency for the spots shovm in figure 156a to spread and join, not in any direction but in definite lines. The figures to the right and above figure 165 illustrate the tendency for the markings to join in the line between the pigmented areas of the media and cubitus and for the individual markings to still retain their characteristic form. On this basis the unusual and aberrant patterns such as 150, 151, 152 and 160, 161, 167, plate XII, are easily explained. In spite of the extreme extension they are like 157. Figures 170 to 187, plate XII, show the patterns of species in which the longitudinal striping lias been developed chiefly- in conjunct- ion with some cross bands, but in which there is no suggestion of the characteristic middle band. Figure 169a .shows the pattern of an Aus- tralian species in which all the dark and light longitudinal stripes are represented. The dark area over the subcosta is clearly distinguish- able. In 169a this subcostal dark stripe is reduced but still present. 427] COLORS OF TIGER BEETLES— SHELFORD 33 Fi^ire 169 shows the extreme extension of the white ; 168 shows a reduced pattern of the same type; 175, a species with three represent- ed simple stripes ; while 183 has only one stripe ; 170 and 170a show the variation in one species in which the middle white stripe may be either present or absent, and the two posterior cross bands are present and curved like the end of the elytron. In 171 the cross band is broken away from the innermost longitudinal stripe in the area of the dark line of the media trachea ; 172 shows a wide middle band with the longitudinal stripe represented only in the anterior portion. Figures 173 and 17-4 show types with connections between an outer, unpig- mented side and the central light stripe in the center. Figures 177. 178, and 171 show a combination of the lateral stripe and the cross band 5.6 ■. 180 to 181a show patterns which may have arisen from types like figure 158 above. Comparing 177, 182, 184, 18-la, and 176, one notes varying lines of oblique connection to which attention was called in figures 78 to 87, plate VII. Figures 188 to 231, plate XIII, show cross bands in the Indian-African-Australian group in which reversed obliqueness of the central band 4 is developed. This obliqueness is rare 1 outside this group except in forms with a well developed sinuate middle V^ / band (e. g. Figs. 292 to 298, PL XV). Figures 188 and 188a show ' ' the well developed cross bands, / and 2.3 being joined at the side ; 189 is similar and 1 and 2.3 are joined obliquely : 190 is similar but reduced. 198 and 199 are similar to 188 but have lost the last cross baud and further reduction in the same direction would result in patterns like 197, 205, and 206. 191 to 196 show a series based on the central white stripe variously broken into spots representing cross bands. 200 to 204o and 213, plate XIII, are a series of related species occurring in India which show an unusual oblique arrangement and combination. 209 an African species belongs to a group with pilosity similar and closely related to the Indian group including 201 to 204a ; it shows the same type of obliqueness in the central marking as in 200. 210 to 212, plate XIII, show further modification of the central band and connec- tion with the oblique humeral curve in the line of the central light space. 220 shows a slightly different trend of similar elements which give the combination in 221 or 220 and 219, depending on the trend taken. 214 to 218 and 222 to 231 show the sim])le patterns of cross bands in which the last and usually the first are missing. 232 to 240 show combinations of markings resembling those just noted in Cicindela, in Therates, Prothyma and Odontochila; compare 232 and 197, plate XIII ; 233 and 206, 234 and 197 ; 235 and 219 ; 236 and 219 ; 237 and 188; 238 and 239 with 210 and 213; and 240 with 188. There are resemblances between patterns in other genera and those in Cicindela. One note-worthy African species (PI. XIV, Fig. 242 — compare 34 ILLINOIS BIOLOGICAL MONOGRAPHS [42g with 209, C oscari) shows the umisual oblique bending of marking which characterized the group noted above. This and oscari are how- ever the only species in which it occurs and the group to which it belongs is similar in pilosity to the Indian groups just described. This particular one stands in closest relation to those shown in plate XII, figures 170, llOa. and 171. It is introduced here because at the outer margin its markings represent 2, 5, and 6 with the almost universal central or fourth absent, except at the innerside where 4 seems to be present and obliquely joined to 5. 241, a and b show a pattern in which 5 and 6 are present while 4 is wanting except for a few small dots. This speciesi appears to show a tendency to double longitudinal lines. 243 shows a second African species in which there is a tendency to double stripes but the central cross band represented at the margin. The patterns show,, in fig\ires 244, 244ff, 245 and plate XIV are of especial interest because the division of the second cross band in those numbered 5 and 4 in the preceding figures are both represented as spots. This is of rare occurrence, the more usual arrangement being like that shown in figure 251. Figures 248 a and 6 show the double longitudinal stripes of an African species, a ease similar to those illustrated above in which one of the types of variation is in the direction of the spreading of the white. Figures 247 and 247« show the joining of such markings as occur in 246 and 259 to make a central longitudinal stripe. Figures 257 to 261, plate XIV, show unusual patterns of spots, which fall into the usual cross bands on the whole, but those in the inner margin of the elytron are usually shifted out of line. Figures 262 to 280 show various directions of reduction of markings in patterns of the type shown in figures 266, 274 and 274o. Those at the left show the loss of the central stripe and those to the right the loss of the inner markings, entirely or in part. 281, 282, 283 show the extensions and obliquity in the type pattern showm. Plate XVI, figures 292 to 306, show the American species in which cross bauds 5 and 6 are separated as seen in 289, 294, 293, etc. The general tendency is for the markings to disappear from the anterior to the posterior end. The component parts of the oblique vitta of some species of the Mexican group is illustrated by figures 311 to 313 and 319 and similar components making a somewhat different vitta in 291, 296a, and 297. Figures 315 to 328 show patterns in which the last or apical (7) cross band is missing or in which variations arise in which it is reduced. Figures 329 to 355, plate XVI, show the species chiefly Eurasian, a few American, in which bands 5 and 6 are present and separate, the former illustrated by a marginal spot behind the center. Figure 347 shows a narrow longitudinal stripe extending forward from the spot 429] COLORS OF TIGER BEETLES— SHELF ORD 35 near the apex, this is an unusual variation in a race of a European species. Figures 361 to 363 show a tendency in certain species for the formation of a vitta in the space between the subcostal and radius (tracheae). Figure 364 shows an unusual joining of the marking of a specimen of C. limbalis loaned by Professor H. F. Wickman, in the space between tlie subeosta and the radius, though the species rarely lias the markings joined and when so not in this line (A) but in line a. 365 shows an aberrant marking in the central part of the elytron of C. campesiris, which is a common European species. 366 shows the darker spots about the white marking iu a closely related species. 370 to 377 show the patterns of other genera ; compare 370 and 362 ; 371, and 185 ; 372 with 367 : 373 with 367. Figures 402 to 478 are presented to show series of unusual com- binations illustrated by the Indo-Australian group of species. 378 shows a marking projecting backward composed of the band 2.^ and the longitudinal part of the pattern plan whicli lies between the media and the cubitus (tracheae) ; the lettered number of the same species shows the extinction of the white. 386 shows an miusual tj'pe of pattern in which the curve appears to rise in cross-band J while the light sti'ipe between the media, and cubitus is obliquely joined in the anterior end to the central spot at the elytral base. Extension of the white is common in variations in this group (383, 384, 385), 379 to 382 show a combination between the middle band and the central basal spot and .spreading of the white. 389 shows a similar pattern but with the joining in the cross-band 2.^ and extension of the white. 387 and 396 are somewhat generalized, representative of the type in question whicli with slight modifications may have led to the 397 and 398 series of patterns (/) or by extension tq the 392-395a series and 400. The balance of the illustrations show the unusual patterns of the Ciciiidelas both reduced to a single marginal stripe and in full form. Most of the species represented are from Australia and New Zealand. Figures 422 to 454, plate XVIII, show^ the imusual marking of Cicindelas with slight distortions, but all the patterns belonging to groups of species which show a strong tendenc}- in the chief representa- tives to varj' in the direction of nearly all white individuals. The irregu- lar and oblique marking in figures 422 and 423, representing two South American species, shows an unusual type of degeneration of the system. The peculiar irregular, branched and scattered character of the mark- ings of several groups shown indicates the breaking up of the system of marking which has been designated as the type upon which they are based. The different species are characterized by peculiar turns forward 36 JLLIXOIS BIOLOGICAL MOXOCRAPHS [430 of certain markings. Compare for example the anterior eross-baud (humeral lunule) of 436 and 427; one is turned forward with a char- acteristic curve, the other backward. This is a difference between the two species which holds good throughout all the individuals. The extension of the white shown is clearly associated with a degeneration of some of the chief tracheal trunks. From this large series of figures we must not permit ourselves to judge that all types of pattern are equally common and equally general in the species of the genus. Figures 329 to 333, and figures 130 and 131 show the commonest and most characteristic types in the genus which are universally distributed and make up vast majority of the grand total for the world. Th|s, the first definitely directed tendency in the group, has been the union of spots to form the characteristic markings of the group shown in figure 49, plate V, as combination of Al, A2, B2, or B3 to make the humeral lunule so called, of A4, B4, and B5 to make the so- called middle hand, and of A6 and/ to make the apical lunule of the taxonomists of the group. If these three types of joining are granted as the first directive principle entering into the make up of the patterns of the group it must also be noted that it does not apply to the majority of species in nine of Horn's groups (XXVII-XXXVI) includ- ing 40 species (Figs. 188 to 215 and 220 to 231, PI. XIII). A few patterns Avith middle band and apical and humeral lunules, and which have three spots in the basal and anal portion of the elytron, are included in these groups and differ from most others of similar components in the presence of these spots (Figs. 273 and 274, PI. XIV, and 163 and 164, PI. XII). These few are the only rei^resentatives which show this char- acteristic middle band humeral audi apical lunule. It applies to only 16 species of the Horn's pilosity groups XVIII to XXII which include 66 species in Africa (Figs. 147 to 149fl, PL XII; 269, PI. XIV; 156, PL XII; 265, 241 to 272, 278 to 280, PL XIV). Of the figures cited, 156 and 265 are of the most primitive type and 266, 267, 275 and 278 show modifications. If we grant the majority of the remaining 500 species show these characteristics as variations or that they may for purposes of discus- sions be assumed to have been derived from forms which did have the three characteristic markings we note that in general the patterns except those mentioned above fall into two parallel series one witliout the spots, including the majority of species, and the other with them, including a' comparatively small number of species. Those with the three spots are confined chiefly to the land directly bordering the Indian Ocean being especially numerous in Africa and India. Spots may be wanting in some variants of such species as cscheri (Figs. 267 and 268) 431] COLORS OF TIGER BEETLES— SHELF ORD 37 aud monteiroi (Figs. 276 and 277). These belong to groups which nor- mally have them, but thej' almost never occur in groups which do not show them in a majoritj^ of members. Considering the compoients of the three spots, the anterior central spot {Bi, Fig. 49, PI. V) is a part of the basal cross baud / clearly shown in figure 179. The anterior one in stripe C, figure 48, plate V, appears to be a fusion of spots Ci and C2 and the posterior one of C3 and C4 as a rule, though sometimes the posterior one is C4 and the anterior one Cl.2, figure 165. There is a tendency indicated by variation to drop oiit these markings in many species. In flexuosa usually C 1.2, i. e., the basal sutural spot, is first to go. In others this is not true as a nde, as shown in 261, 276, 277 and 280. On the other hand there is no species in which these are present and other markings absent. These facts indicate that these spots show a tendency to disappear first, leaving the types of pattern without them, ilore rarely they may unite to form a band which may persist in the extremely modified forms, figures 151, 160, and 167. One of the characteristic types of marking which seems to belong to almost the entire group showing the typical middle band, is the oblique shift- ing of the cross band which makes the humeral lunule. The tendency toward obliqueness of the middle band of the typical forms seems quite general in many groups but by no means iiniversal, and is sho\^m by some species in all the groups, and hence is illustrated in all the groups of figures : 157, 163, 222. 227, 273, 276. 288, 299, 451, 33.5, 336, 342, 411, and 417. In other groups another tendency seems to be present, namely to- ward a sharp forward-bent angle on the middle band (Fig. 482) figures 209, 206, and many others in which the usual combinations have not been affected are sho^vn in plate XIII. On the other hand scarcely a species in plate XII shows this tendency except figure 150. Figures 292 and 293, plate XV, 339, plate XVI^i and others related show the same tendencj'. It is shown in the patterns of the Australian group (Figs. 394 to 396, PI. XVII) where a middle band involving different elements occurs, and is particularly conspicuous and characteristic in some of the Mexican and South American species (Figs. 428 to 434, PI. Xr\"III) where it is the chief distinguishing feature. In the, group as a whole the most striking tendency is for the markings to disappear, beginning in the proximal anal region of the elytron and usually leaving the more posterior distal markings present. But to this tliere are many exceptions in which the central marking on the elytron is the only one left. (See figures 255, plate XIV; 222 to 231, plate XIII; and 206.) Another tendency manifested in many species is the extension of the white; it is seen to crop out in all groups from any starting point which is in existence and to proceed from tlie spots characteristic of 38 ILLIXOIS BIOLOGICAL MONOGRAPHS [432 the group, in the direction of general concentric extension in wliich the original type of pattern may be recognized (Figs. 160, 167, 169, and 181, PL XII ; 204, 204a, 196, PI.'XIII ; 378 to 437, Pis. XVII and XVIII) . Thus one who inspects the figures as arranged is inijiressed with the fact that there are a great many directions in which patterns have been modified and these figures are numerous and intentionally substi- tuted for less satisfactory descriptions. The material afforded by the 600 or more species is rich iu possibilities and excels in this respect the butterflies of Eimer or pigeons of Whitman. EXPERIMENTAL MODIFICATION OP PATTERNS To test the laws of modification of the typical patterns of Cicindela larvae of several species — C. tranqucharica, rcpanda, hirticollis, lim- balisylepida, and lecontci — were subjected to low temperature, high tem- perature, and moist and dry conditions. The temperature was raised about 10 degrees C. above that encountered in the normal outdoor life history. The experiments were carried on in the apparatus sho^^^l in figure 455, plate XXIX, and described in connection therewith. The larvae were put into the high-temperature (near 37°, 1906; 40°C, 1905) about May 15. They were placed in a lamp chimney con- taining fine sand. The apparatus as arranged gave 2° to 4°C. higher temperature at the top than at the bottom. The average of the two was used in computing the mean. Temperatures were taken twice a day as a rule. The temperature rose each day as the sun shone on the eases so that during tlie hottest weather daily maxima in soil temper- ature went to 40 to 42 degrees at times. The results of the experiments on C. tranqucbarica so far as the patterns are concerned are shown in figure 456 a to g, plate XX, and 457 ato&,458,459 and 460;these should be compared with control 456 o' to &', w', 457 a' to e', w' and 458 a'h'. A comparison of these experiments with their control and the representative of the forms collected in the field from the same generation shows that in tlie controls the normal middle band reaches to the margin of the elytron where it is expanded in the line of the longitudinal band a A; the longitudinal part is parallel with the anal side of the elytron ; the middle band is hooked at the end or turns into a horizontal position in compliance with the normal direction of the transverse band from which it is derived. The humeral lunule is usually hooked. The angle iu the middle band is a right angle and there is a forward extension of the middle band at the angle. The patterns which result from tlie experimental conditions almost without exeejition differ from the control in the following respects: 433] COLORS OF TIGER BEETLES-SHELFORD 39 1. The humeral lumile is usually without any enlargement at the end suggesting au expansion in the place of spot Bj and j. 2. The middle baud is withdrawn from the margin in all cases and in only one case, figure 556 g, is there auy longitudinal extension. 3. The angle of the middle band is always less acute and the for- ward extension less pronounced. 4. The longitudinal portion of the middle band is oblique to the anal or inner margin (suture) of the elytron. 5. The end of the middle baud is not hooked but rounded, and rarely even parallel with the transverse bands. A close examination of the marking of the experimental individuals show that there is correlation in all the respects in which the middle baud is modified, in general the most oblique middle band is almost withdrawn from the margin and shows least hook at the end. Figure 461 shows au unusual type of marking and of modification, the most reduced marking in specimens of C. linibalis subjected to the same experimental conditions as the tranquebarica shown above. The usual type of modification which is quite general in experimental speci- mens has the longitudinal portion of the middle band shortened. It is also more oblique and thus less like the simple type. The middle bands of these specimens approach those of the variety splcndida (Kansas). They represent a more extreme modification of the simple type than the experimental middle bands of specimens of C. tranquebarica. The markings in two out of about twenty individuals (Fig. 461) surviving the liigh temperature showed a sharp bend for\\ard. This is the reverse of the usual tendency in the purpurea group but is a strong tendency in some other species shown in plates XIII and XV. One indi^^dual out of several hundred collected from the habitat in question, reared as controls, and reared for ontogeny showed this character. Apparently the tendency to respond by a sharp forward bend is little developed in purpurea. Figure 463 a to d, 464 a to c, and 466 show the patterns resulting from the high temperature experiment witli C. Iccontei while 466 a' to c' and 467 u'', x', y', and z' show the conti'ol which survived and the range of variation in a series of specimens collected from the same area from which the larvae for the experiments were obtained. First of all the high temperature experiments show patterns with reduced markings. The markings shown in 463r; are joined in a way which rarely or never occurs in the stock from which they were collected and which is on the other hand characteristic of the varieties of this species which occur on the Atlantic coast. Also 463rf shows a pattern which is smaller in markings than auy that have ever been collected near Chicago, 40 ILLI.XOIS BIOLOGICAL MOXOGRAPHS [434 467 II representing the smallest, ■which makes the marking of 463 un- doubtedly reduced by experimental condition. Figures 465 a to 6 show experiments in which the larvae, pupae were iced from the beginning of the pupal stage; all either by remark- able accident or through the effects of the experimental conditions show the widest type of markings; a third specimen was only slightly modi- fied. In 465& the form of the end of the elytron is rounded in an unusual way and the surface appearance of the entire body and the elytron are different from the normal types. Figure 46Sa and w' show the type of modification occurring in experiments on C. hirticoUis. The middle band is modified as follows: the hooks aud angles are rounded, the transverse part which usually turns forward and has a sharp angle as in 468 w' is oblique in the opposite direction. These modified patterns are identical with those in southern and western localities. This modification is of the same kind as that in C tranqueharica and C. purpurea. Thus it is evident that C. tranqueharica, hirticoUis, and lecontei may be modified in structure and pattern by high temperature during the pupal and prepupal stages. Experiments performed on C. repanda, Icpida, and punctidata show no such modification, or pattern modifica- tion of any other type so far as has been noted. Specimens stimulated by al temperature of 37° C. in the fall and forced through the winter were modified only in case of the specimens which emerged early, January 1. Specimens which emerged in the spring earlier than the normal were not modified. One specimen of C. hirticoUis (Fig. 566, PL XXXI) coming through without any winter was very much smaller than the normal. A specimen of C. lecontei shown in color plate XXIX, figure 556, was different in form, the abdomen being broadest at a point not usual for lecontei. One of the patterns of tranqueharica produced in this waj* (Fig. 459) was one of the most striking modifications obtained. Thus so far as the species which show modification are concerned the modification appears to be in definite directions and the modifications of C. tranqueharica, C. hirticoUis, and C. limhalis are in the general direction in which the modification of the pattern plan has proceeded in many patterns which have deviated from it in course of their evolu- tion. The experimental results further show a basis for the interpre- tation of the geographic variation of the group which is our next topic for consideration. GEOGR.\PHIC VARIATION OF PATTERNS C. tranqueharica, very widely distributed in North America, (PI. XXII) shows great variation in color and markings, but the 435] COLORS OF TIGER BEETLES— SHELFORD 41 extreme forms are comparatively rare and eoufiiied to the Pacific states. Plate XXI shows the classes into which the patterns of this species may be divided and their distribution. The graphs represent the distribu- tion of the per cent of classes shown by the figures below for specific localities. It will be noted that tj^pes g and h which correspond in middle band characters occiu" occasionally as extremes especially in Kansas and Texas localities, while west of the rockies where the summer and springs are dry and favor high soil temperatures these tyY)es are fairly common. This type of marking with middle band reduced at the margin makes up a considerable percentage of the individuals collected at Hagermau, Idaho ; San Bernardino, California ; Provo, Utah ; and Las Vegas, Nevada ; but they are nowhere the dominant type. In certain Nevada localities the retirement of the middle band appears to begin at the inner end and the withdrawal from the margin follows only in very reduced types. The type with the middle baud with- drawn occurs in southern and western localities. Twelve per cent of the specimens from central Texas show middle bands like those modified in experiment. On the whole there is a correspondence between high soil temperature and the reduced tj'pe of markings which accords with the experimental results. Plate XXIII shows the geographic variation of C. scutellaris and its varieties ranked as aberrations by Horn. The series of classes shown beginning at the extreme left are from the northern portion of its range in New England; passing to the right are shown verj- reduced markings at Raleigh, and very rarely any markings at all at Mobile and in Texas localities or points in western and west central states: Oklahoma, Kansas, Nebraska, and South Dakota, Colorado and New ilexico. In all localities, however, on and east of the Missouri River in the central states, there is a noticeable increasing in the size of mark- ings as we pass to more northerly localities and to more easterly local- ities as fai* as Chicago. East of Chicago the marking of specimens from along the lake shores are not larger than those taken at the south end of Lake Michigan. As will be seen from the graphs (PI. XXIII) the range of variation is least in the gulf states localities where the markings are most reduced. There is further a noteworthy difference in the Mississippi Valley and Atlantic Coast forms. The humeral dot {ai, Fig. 48, PI. V) is never present and the so-called posthumeral dot {A3.3, Fig. 48, PI. V) is seldom so except in the more northern localities and is never large when present. It is never joined to the middle band {A4, B4). The markings are massed in the posterior half of the elytron on the costal margin. In the forms from Missouri River localities and eastward the humeral dot is usually present — always present in the more eastern 42 ILLINOIS BIOLOGICAL MONOGRAPHS [436 form — and its absence is associated with extreme reduction of the markings in general. Thus patterns made up of a row of dots on the costal side of the elytron are the most numeroiis in Iowa localities and probably those just east of the Missouri River. Thus the selected classes of individuals are geographic in their relations and hence true classes. Further evidence for this statement is shown in plate XXXIV where the color differences are indicated, showing that the immaculate forms are further divided into races on the basis of color. Those of the humid southern states are green, and those of the western steppe, with its dry early summer following early spring rains, are red. In full accord with the experimental results cited above are certain differences in patterns of two localities from which collections were made often. The larvae used^ in experiments were collected from a point just north of the village of Miller, Indiana, from a small area of oak dimes about an acre in extent. Adults were collected from this same locality dviring several years at various times in the season and differences in color and pattern were noted. Graph 10 is the distribu- tion of classes in 200 individuals belonging to the generations of 1904 and 1905. This same graph is repeated above on a smaller scale with graph 11 added, which shows the distribution of classes in .51 speci- mens collected from the same area in April, 1906. Graph 12 shows the distribution of classes in a series of 60 specimens collected in the north- western part of Gary, (600 ft.) (Pine Station, Indiana,) in April, 1906, showing the modal class to be o instead of q and a small percent- age of individuals with markings joined. Graph 13 shows the distribu- tion of classes in a series of 37 specimens collected in September, 190S, in which the same difference is shown. A difference in the distribution of classes is indicated by a comparison of Graphs 12 and 13. These differences are striking for one who is familiar with them. The dif- ferences between the Gary and the Miller locality were noted while collecting the species in the two localities during several years. The specimens collected in Gary showed those with markings joined as very rare. The entire series from the Gary locality show the same thing. There are also similar differences from generation to generation, in the catches from Miller. The difference in the conditions at Miller and in the Gary locality is striking particularly during the larval and pupal periods. The area in Gary is covered with scattered pines and in places from which some of the specimens were collected cottonwoods occur. The area is one of lake sand on which cottonwoods grow up and are succeeded by pines and the pines by oaks. The Miller locality is an oak dune area with well-established growth of oaks. One mile south of the Gary locality are oak covered ridges. Specimens from jiere are of the usual type taken in the Miller locality. Many of the 437] COLORS OF TIGER BEETLES—SHELFORD 43 nines had been cut off the pine belt in Gary wliere my specimens were collected. It is about as open as the cottonwood belt where evaporation from the porous cup atmometer is about twice that of the oak dunes in which the ]Miller specimens were collected. The soil temperature goes very high in the Gary locality. Distance below Temperature in degrees G surface Air 36 C 1 1-4 cm. 47 3-4 cm. 38 8-9 em. 35 10-11 cm. 33 12-13 cm. 32 17-18 cm. 30 These forms pupate at a depth of 15 cm. and thvis at a tempera- ture of 31°C. on the warmest days. The temperatures in the shade in oak covered sand dunes are much lower being about 27°C. under the same conditions. Plate XXV shows the division of the various subspecies of C. pifr- purca into classes. Here the primary division of the group, shown in the immaculate form in the center of the group which is ver^v rare, is an habitudinal one — those at tlie left are the patterns of a series of races which inhabit level ground usually among scattered vegetation. To tlie right are tliose that occupy steep banks, particularly clay banks. Classes a and b, cimarrona, and t, 10 guttata, do not appear to be so differentiated and accordingly the graph perhaps should have been revei-sed with the generalized patterns in the center, though further investigation would be necessary to determine this. The present arrangement is based on resemblances between the two, cimarrona and those at the left, and C. 10 guttata and those at the right. The distribution of the two gi-oups shown at the right and the left of the center are shown in figures 471(7 and 472. If one notes the localities represented by the graplis showing the distribution of classes, it is evident that there is no striking difference in tlie distribution of classes in Puget Sound, Massachusetts, and Color- ado. The modal ela.ss for Manitoba, Topeka, and Chicago, is tlie same. This goes to indicate that the main line of separation is habitudinal rather than geographic. Similar relation could be shown for other species. The main differ- ences in patterns are primarily associated either with different local- ities usually separated geograpliically, or witli differences in habitat preference. The figures on plate XXVI II (Figs. 473 to 536) are arranged 44 ILLIXOIS BIOLOGICAL MONOGRAPHS [438 in parallel lines of similar patterns. Thus figures 473 to 485 are pat- terns of C. iranqucharica similar to those sho^^-n in figures 486 to 494, exceiJting 481 and 483 which are diiferent species closely related to C. tranquebarica. In figures 486 to 490 are shown a series representing the typical patterns in C. scutellaris; it will be noted that these parallel tliose of C. tranquebarica with most reduced markings. Also figures 491 to 496 show the pattern of tlie Great Basin group of species and varieties to which C. fulgida is closely related. These parallel some of the patterns of C. tranquebarica and are in turn paralleled by those of other species. Concentric extension of the white likewise character- izes the patterns of the group. Figures 497 to 501 show a series of patterns in C.pulchra which are roughly parallel to tliose of C. tranque- barica and very closely parallel to those of C. scutellaris. The com- monest pattern of this species is, however, figure 498; 499 and 501 being rare and collected only near Alpine, Texas. Figures 503 to 505 show the series of patterns of C. longilabris which parallel the patterns of other species shown above and below. Figures 506 to 518 show a remarkable and long series of patterns of purpurea' parelleling the entire tranquebarica series without the addition of other species. The entire series is however different than the other series especially different from the tranquebarica series be- cause of the short humeral lunule which always stops with spot A2.3 while that of C. tranquebarica is made up of A2 and B3 in oblique combination (see Fig. 49, PL V). Figures 522 to 527 show the markings of the C. sexguttata group which parallel those of the other groups quite well throughout a series of five types. Figures 528 to 536 show a series of types belonging to five closely related species. The patterns at the extreme right show extension of the white which appears to have occur- red as a tendency taken at an,y point in the series represented; thus figures 520 and 521 belong with 488 and to the same species. Figure 519 belongs with 531 and represents a different type of extension. While a general parallelism is shown by the series of patterns, there is also a characteristic series of small differences belonging to the usual types of most species. This indicates that specific cliaracters in the color patterns are matters of detail and am^ definitely directed specific or racial tendencies woiild have to be based on a consideration of such details rather than the general plan of the pattern and the general parallelism shown in the group of figures just discussed. While specific patterns are often very closely parallel, one who is very familiar with them can identify the species from a single elytral pattern in the vast majority of cases. Considering the pattern of the rest of the group, represented in figures 473 to 537, C. formosa and its varieties is distributed on the 439] COLORS OF TIGER BEETLES— SHELFORD 45 Atlantic coast and foi- some distance inland in llassaehusetts to Mary- land where the markings are of the tj-pe shown in figure 532 and slight- ly wider with the all joined at the side. The sharp forward bend of the middle band is characteristic of the eastern forms. C. formosa is distributed about the sand dunes of Lakes Michigan and Erie and through the sand areas of the central states, the distribution being very nearly like that of C. sciitellaris except that formosa is wanting from Virginia to Texas along the Atlantic and Gulf Coasts. The markings of the western Mississippi basin forms are broad as shown in figure 531, plat* XXVIII, while m the more southern and western forms from Texas, Colorado, and Oklahoma are characterized by a middle band tending to be straight across the elytron. The species which stands close to this is C. venusta (Figs. 533 and 534). The pattern is similar to that of C. gencrosa. It occurs only in sand areas of the great plains. The southern repi'esentatives have markings similar to figure 531 in width, but in Manitoba there is a tendency to the extension of the white as shown in figure 534. G. lim- hata is a closely related species which is taken only in blowouts in sand hills of the western Nebraska region and of Manitoba. Figures 535 and 536 show typical patterns. They do not vary greatly geogi'aphically. C. ancosisconensis and duodecemguttata are invariable species (Figs. 528 and 529), rcpanda a subspecies of 12 guttata distributed almost everywhere east of the Rocky mountains in the United States and Can- ada. Specimens from Louisana, Manitoba, and Virginia do not varj' appreciably. The lan'ae inhabit very moist soil and soil temperature cannot be of any magnitude. The habitat and larval habits are such that variations due to differences in temperature and moisture are not common. If the soil becomes too dry the larvae leave it and dig a new burrow in soil of the wetness required by the species. Since they occur near water courses, this tends to keep larvae in similar conditions no matter in what latitude they occur. The variation of oregona. a related species, has not been studied. C. hirticollis occurs on the sandy shores of the sea, lakes, and rivers from Vera Cruz to California, the Great Lakes, and Massachu- setts. The pattern which is shown in figure 330, plate XVI, is quite invariable as compared with the rest of the species considered. High temperature experiments performed with these showed clearly recogniz- able modification in which the pattern duplicated Southern and South- western forms. The experiments and geographic and otlier variation are likewise parallel. C. scxguitata has been studied and shows peculiar variations. Spec- imens from the Northeastern United States and the region of the Great Lakes have well developed markings (Figs. 525 and 526, PI. XXVIII). 46 ILLIXOIS BIOLOGICAL MONOGRAPHS [440 The same is true of Texas specimens. Specimens from E. Tennessee are reduced as in figure 523 and those from eastern Kansas are usually immaculate with a few like 523. C. punctul n o X X X X X X X X X z X X X X X X X X** X X X X* X X > > > X X X X X X X X X X x X X X X X X X X X X X X X X*** X X * tranquebarica nigrocoerulea oregona formosa hirticollis repanda willistoni fulgida pulchra X anthracina X X X X X X X < X X X X X X X < cuprascens lepida X *Reflections in western forms. **Reddish brown. ***Dull bluish drab. Vera Cruz, Me.xico. In scutellaris, purpurea, anthracina, and sexguttaia black and green forms are mixed, i.e., the species are dimorphic. The same is probably true of tranqueharka, as plutonica appears to be rare and 445] COLORS OF TIGER BEETLES— SHELFORD 51 occurs in California where the usual population is green. The physio- logical condition in which no metallic film is secreted is closely related to one in which a metallic film producing green is secreted. The secretion of a film which lies at the outside of the primary cuticula in the first work of the hypodermal cells. It would seem that the secretion of such a layer might be inhibited by environic stimuli! at a critical stage in the life of the pupa, but there appears to be no experimental results showing whether or not this is true. If environmental conditions do influence the occurrence of black and green, climatic conditions applicable to all species are not alike (see p. 52). In the case of C. scutellaris the green and black forms have least pigment developed in the elytra (black is accompanied by a similar amount), and green in ontogeny is accompanied by least. The amount increases as the reddish color comes in, in lecantei. The amount of pigment in the brilliant red western form is intermediate between the green form and the dark red lecontei. C. splendida, very brilliant, shows much less pigment than limbalis, which is dull. Many species, particularly purpurea and pulchra, show more bril- liant colors along the elytra margin where white markings usually occur. This is noticeably true in purpurea, which in the subspecies cimarrona has a complete white margin in many specimens. As a rule when the areas commonly occupied by markings become pigmented the colors in these areas are more brilliant. W. Horn (1915) has called attention to this. As has been noted, the elytral surface of most tiger beetles is made up of small hexagonal pits which jirobably convspond to the hypodermal cells which secreted it (Fig. 1, PI. I). The ridges between these lie over the boundaries of the cells. In the elytra of C. purpurea these pits are smaller in the blue-green margin. The same is true of many other species as shown in Table III. Wliile many colors such as green and greenish blue, red, etc., in early ontogeny change to colors of longer wave length during ontogeny and later life, such is not true during ontogeny at least in the case of such purple specimens of C scutellaris. These are rare and only a fe\v specimens from Starved Kock (Utica), Illinois, have been found; .some of these are purplish brown, but one individual was secured in the larval stage and reared (Fig. 558, PI. XXIX). It was purple from the beginning and, never showed any tendency to change, though it was kept for a long time. The same is probably true of the purple forms of C. sexguttata which occur in eastern Kansas; purple forms of nigrocoerulea show no blends with the green. tJ? I 52 ILLINOIS BIOLOGICAL MONOGRAPHS [446 Table III The following table shows the relative size of hexagonal cups in various forms and parts of the same elytron, etc. Species Variety Locality Organ Part Color Diameter in mm. C. purpurea.. " scutellaris.. " generosa.. chinensis.. deni)erensis. lecontei scutellaris rugifrons modesta ■— Massachusetts... elytron Chicago Penver . Chicago.. Colorado China margm disc margin disc disc green., red green.. brown.. f red i " [ green., green., black.. red \ blue [ metallic 0.0115 0.0150 0.0115 0.015 0.0150 0.0150 0.013 0.0115 o.oioo 0.0 1 00 0.0150 f 0.015 to ] 0.0225 [ 0.018 av. 0.013 GEOGRAPHIC VAEIATION IN COLOR The black forms of C. scutellaris are found to occur in some New York localities, and some New England localities, but are less numerous than green ones. A complete catch from Providence, Ehode Island, for one season, including hibernated and freshly emerged forms, showed less than 20 per cent black individuals ; a similar catch from Framingham, Massachusetts, gave no black individuals; 112 specimens from Aqueduct, New York, showed about 15 per cent black. Some localities in New Jersey show, according to Leng, a majority of black forms in spring, A small catch from Baltimore, Maryland, showed more than half black forms. At Raleigh, North Carolina, black forms do not occur, and I find no records for Virginia, North Carolina, and South Carolina ; but black forms occur in Alabama, Georgia, and Florida. At Mobile a few black ones are found in the autumn but very few or none at all in the spring, according to Messrs. Loding and Van AUer who have been interested in them for several years. None are recorded for points farther west. Black forms of C. sexguttata likewise occur in the eastern states. New Jersey and Pennsylvania, but not in the southern localities. Black forms of C. purpurea (see map, Fig. 472) occur in Illinois, Iowa 447] COLORS OF TIGER BEETLES— SIIELFORD 53 Minnesota, Kansas, Nebraska, South Dakota, Colorado, Utah, Wyom- ing, and New ilesico, but are very rare in eastern localities. Mr. C. A. Frost secured one bluish black individual in Massachusetts. The black forms of C. tranquebarica are recorded from a single locality in California. No black forms occur in the localities where black forms of other species occur though blackish green forms occur in the Pacific States and blackish brown, in the Gulf States. Likewise there is no correlation between geographic conditions and green forms. Scutel- laris is green on the Atlantic coast, purpurea in the central and north- ern great plains, tranquebarica on the coasts and coastal mountains. Exclusive of black forms which have just been discussed the geo- graphic variation of colors in the species belonging to the tranquebarica group, may be stated as follows: Geographic variations in color are of special interest in the case of C. scutcllaris; I note green forms pre- dominating in all specimens in the Atlantic Coast and Gulf States. Bluish retlections characterize these as a rule, particularly in some localities where occasional blue forms occur (Fig. 470 a). In Texas along the northeastern border near Oklahoma forms occur with a decided golden cast which in series in some localities range from bluish green through green with golden cast to flame red like figure 554, plate XXIX ; north of this flame red predominates. Forms with flame red eh'tra and green or blue thorax occur west to the Rio Grande, occupying a triangular area with its apex just north of the Black Hills and eastern point near Topeka, Kansas. Points a short distance west of the Missouri River such as Topeka, Kansas, and Su- perior, Nebraska, show great variation in marking and all intermedi- ate color conditions between the forms with flame red elytra and those of tlie dull brown and wine color occurring to the east and north of the ilissouri River. The most brilliant wine colors occur between the Mississippi and Jlissouri Rivers and in Manitoba; near Chicago the brilliant wine colors are not usual, but greenish browns and green- ish individuals are common. There appears to be no close correlation between the distribution of these colors and any mapped distribution of factors. C. purpurea is very variable ; figures 471 a, 472 show color varie- ties of this species. In general among the groups in which tlic mark- ings are withdrawn from the margin, the forms with the n[)per part of the elytron reddish and its margins green are most widely distrib- uted, extending almost throughout the range of the species except the Pacific coast specimens which are golden green (Puget Sound, 10 ft.). / The eastern forms are of the typical red elytroned type. In the entire "/ ^Mississippi Basin, Great Plains, and Salt Lake Valley tliis is mixed with green and black fonns, the latter two predominating in the west- J. 54 ILLfXOIS BIOLOGICAL MONOGRAPHS [448 ern Great Plains. In the New Mexico localities dark brown forms (ciniarrona) occur. There is no correlation between color and mapped climatic conditions unless it be rainfall. Considering the purpureas in which the reduction of markings leaves only a small dash at the margin of the elytron, one notes that the wine colored specimens are distributed throughout the region, of the Great Lakes and in Manitoba and generally westward to the Mis- souri River, .and Colorado. This type is distributed in a general way north of about 41 degrees North Latitude and has the thorax the same color as the elytron. The forms splendida and transversa are similar in color but have the thorax green or blue and the elytron either red or wine color, they are distributed south of the form with red thorax and in the eastern part of the range ai-e less brilliant than farther west. The more western forms have brilliant red elytra similar in color to that of the red scutellaris. Mixed with these are the green forms; in western Kansas and Colorado, especially, they occur with the red forms and are often taken in coitus with them. The green form is evidently merely a color aberration of the red form. The color variation of C. tranqueharica is not striking over the entire area east of the Rockies. Nearly all are simply dull brown. Specimens from the moist southern states are usually duller blackish brown than the northern forms. No striking color varieties occur even east of the Pacific states and Idaho. In some parts of eastern Califor- nia (Bridgeport) they are brown, while only a little waj^west they are green ; further surprising differences were found in Nevada. At Caliente the writer took brown tranqueharica and blue oregona, while nt Las Vegas he took green and bluish tranqueharica and no oregona, ^ . hich occur there and are probably green also, but there is no apparent reason why oregona should be blue or green and tranqueharica brown in a region where both are likely to be green. C. generosa is brown and wine color in eastern localities and where purpurea is similarly colored. Near Chicago the colors are simi- lar. At Topeka, Kansas, the color varies considerably, reddish, bluish, and greenish brown occur. South, and sotithwest from this point the specimens are progressively redder. The most brilliant forms are the red ones from western Oklahoma, western Texas, and Colorado. At low altitudes these are golden red. Wine red occurs at high altitude (Salida, Colorado, 7,000 ft.). C. hirtieoUis has already been discussed (see page 49). With the exception noted there is little variation and distribution is transcontinental and from the Great Lakes to Vera Cruz. 449] COLORS OF TIGER BEETLES— SHELFORD 55 EXPERIMENTAL MODIFICATION OF COLOR This is froiight by many difficulties on account of the remarkable series of colors and color changes occurring in ontogeny, and the usual early death of individuals reared under experimental conditions. Fig- ure 555, plate XXIX, shows an experimentally modified individual of C. lecotitei. The presence of the yellowish color in the markings indicates that secondary cuticula has been secreted with the air spaces between, in quantity sufficient to give the opaque appearance to the markings. This specimen in particular was known to have died 15 days after it was dug out of the soil, which is not until the cuticula is well hardened. Its markings are reduced below anything ever found near Chicago. The color shows an unusual amount of yellow and approaches most nearly to some of the western forms of scutcllaris (Fig. 554) though not exactly like any forms known to occur. This particular individual showed more yellow and was most generally modi- fied, leaving no doubt as to the fact that color modification had oc- curred. Three other individuals, all of wliich lived long enough to show the development of opaqueness in the wliite markings, were pro- duced and showed green of unusual clearness from reddish brown and suggestive of green forms rather than the parent stock of Iccontei. All these were in dry conditions. The warm moist experiments show-ed green forms but not clearly differentiated from ontogeny stages in part due to earl}' death. Three specimens (Fig. 557) were brought through successfully in icing experiments and lived two weeks or more. Two of these were characterized by broad markings and dull brown elytra and rather striking differences between the color of the head and the thorax, the latter being quite green. Figure 557 shows considerable modification of form and size not noted in the other two. The very rounded ends of the elytra, and square shouldered character was quite striking and in direct opposition to the usual tendency sliown in the rest of the group. Figure 556 shows a specimen brought tlirougli at 37°C. with raarked acceleration of development. This individual was small, slen- der in the head and thoracic region, with the el.ytron widest in the region behind the middle band. The color is much brighter and freer from dull brown reflections than that of the normal specimens, having a decided brilliancy to the color. This specimen was kept alive until the opaque appearance of the markings was well developed. This body form is characteristic of many specimens from the extreme southern states. There is a noticeable general tendency toward this general body form in all individuals reared in high temperature. 56 ILLIXOIS BIOLOGICAL MOXOGRAPHS [450 Experiments were performed on C. hirticollis which paralleled those noted on C. scutellaris, but with results on markings and none so far as color is concerned. It is probable that the experimental indi- viduals showed more green than others, but the difference is too slight to justify an unqualified statement to that effect. One striking result was obtained in the experiments where the temperature of about 37°C. was maintained on larvae which had not hibernated; one small indi- vidual was obtained (Fig. 566) which however retained all the striking characteristics of the species. Experiments on C. tranqucharica were successful. Specimens reared in temperature of 37°C. and much moisture (Fig. 570) showed the dull blackish brown which characterizes the colors of some of the specimens from the moist southern states. This color was not iiniform throughout the series so raised, but was much commoner than in the ease of specimens reared in hot dry conditions, as these are more bril- liant (Fig. 569). A number of specimens were iced but only one of these was especially peeiiliar (Fig. 568). This was decidedly more red than any others seen in the course of my studies. Some of the iced specimens were unusually dull, however, and no uniform results were noted except that the heads were uniformly greener. C. limhalis was subjected to high temperature. In the moist con- ditions dull colors were obtained. Figure 577 shows one of the high temperature individuals in which the color is deeper red and the re- flections more striking blue than in the normal specimen at this stage (Fig. 575). 579 which shows an individiial subject to high temper- ature in moist conditions is morei generally dull green. 578 shows an iced specimen which is similar to the warm moist individual. These differences are slight and not very convincing, but the individ- uals are different from any reared or collected under other conditions. Experiments of a similar character were performed on C. punc- tulata but appeared to be without results. A similar series on C. lepida were likewise witliout results. RELATION OP COLORS AND COLOR PATTERNS TO CLIMATE After a thorough study of the subject and comparison of the dis- tribiitiou maps of several species with maps showing the rate of evapo- ration of water for the year, the evaporation of water from the porous cup atmometer from April to September, the ratio of rainfall to evapo- ration, mean annual temperatiire, temperature April to September, and with maps showing cloudiness, humidity, rainfall, etc., it was demonstrated that the distribution of color varieties, and pattern varieties even where the types are quite distinct, is not correlated with the conditions shown on such maps. 451] COLORS OF TIGER BEETLES— SHELFORD 57 111 general such correlation is closest in relation to rainfall, but this correlation is not so good as one would expect (Fig. 470 a, PI. XXIV). This is perhaps to be expected in the ease of species which belong to local conditions which is true of most of the species of Cicindela. This subject has been discussed in some detail ( Shelf ord, 1911). Here it was shown that species which were distributed in a major climatic habitat had a distribution correlated with the distribution of vegeta- tion, which in turn is correlated with the distribution of climatic con- ditions. I showed further that species such as C. tranqucharica trav- ersed almost the entire continent without much variation by virtue of living in moist soil, due either to climatic moisture or to local stream moisture or lake-shore moisture. C. scutcllaris, C. purpurea and most of the other species noted are found in some special kind of soil such as sand containing a little humus (Shelford, 1911, 1913b) or steep clay- banks or some other restricted situation. Taking C. scutellaris for example, this species being found in well drained or dry sand contain- ing a little humus and bound by scattered vegetation throughout its range, it is to be expected that the distribution of the species will be correlated with some sort of measured soil conditions such as soil tem- perature, soil wilting coefficient, or the like ; but no such conditions have been recorded or mapped. There is some evidence of soil effects in this species (see Fig. 558, PL XXIX). Some specimens from the verj' coarse sands resiilting from the weathering of St. Peter's sand stone, near Utica (Starved Rock), Illinois, are purple. No purple forms have been taken elsewhere. Two specimens from sandy clay (Suman, Indiana) had an unusual silky appearance. When soil tem- perature work under way is published, I shall attempt to make use of the extensive records which have been accumulated for the purpose of working out correlation between conditions and color and pattern va- rieties. Conditions associated with altitude influence color in some cases, but there is no unity of conditions or colors. GEOGBitPHlC CENTER OP THE GROUP ON THE BASIS OP PATTERNS The usual criteria for the center of distribution (Adams, 1902) indicate that the Oriental region or at most the Oriental and Ethio- pian regions (shores of the Indian Ocean) are the geographic center or center of distribution of the group. The first evidence presented which indicates this is found in table I, in which eleven groups of species are shown to occur in the Oriental region and in other regions, while not more than six occur in any one other region and at the same time in still others. Patterns are divisible into three great gro\ips: first those without the spots at the base and along the inner border of the elytron shown to the left of the bottom of figure 580; these patterns represent the 58 ILLIXOIS BIOLOGICAL MOKOCRAPHS [452 usual tyj)e of the group aud are world wide in distribution. The patterns to the right of these are those with the basal spot and the two spots along the inner border, shown on the map by the stippled area ; this includes a number of pilosity groups and thus represents considerable diversity. The group in which the middle cross band (4) is oblique in the reverse direction as compared with that which is usual in the group as a whole, is shown by small circles. This is essentially confined to the Oriental region. There are a few species in Africa which show this and which appear somewhat related on the basis of pilosity, but circles are omitted. The group of species and patterns shown at the extreme right and represented on the map by the short oblique lines constitute a group divided between the Oriental and Ai;stralian regions. An over-lapping of the various t\T3es in the Oriental region is evi- dent. This would place the center for the group in that region but several African species appear to be most primitive from the standpoint of kind of patterns shown. It accordingly seems best to consider that the lands adjoining the Indian Ocean constitute the center of distri- bution of the group. GENERAL DISCUSSION The evidence which must support any conclusions drawn is of such a character and drawn from so many sources that the presentation of a few lines of evidence and the conclusions forthcoming from them can best follow the general presentation of data and minor conclusions on the preceding pages. Since color and color pattern are quite dis- tinct so far as laws governing them are concerned, the discussion of the two will be separated. PATTERN TENDENCIES Under this head we are concerned with (a) the original type, (b) the most characteristic elements and combination of original charac- ters, (c) general laws of pattern modification applicable to groups of species, (d) laws applicable to particular species, and (e) laws appli- cable to subdivisions of species. As has been noted the number of directions in which modification has preceded are numerous and any statement of siich directions is difficult aud has led other authors to make general statements regarding the modification of patterns which were general enough to apply to a large number of species. The earliest account of variation in the color, or markings or the patterns of tiger beetles is that of Dr. Geo. H. Horn (1892). He took 453] COLORS OF TIGER BEETLES— SHELFORD 59 the marking of C. tranquebarica Herbst as the underlying type "from which all forms observed in our Cicindelas have been derived". He bases this statement on the fact that it is the so-called humeral lunule, middle baud, aud apical lunule which give similarity to the patterns of the genus. He states that modification occurs in any one of four ways : A. By progressive spreading of the white. B. By gradual thinning or absorption of the white. C. By fragmentation of the markings. D. By linear supplementary extension of the white. These tendencies are all recognizable, all of them occurring in the course of individual and geographic variation of single variable species. Walther Horn (190S) in Genera Insectorum discussed the patterns from a somewhat different point of view. He states that in the ideal sense the markings which he recognizes as the humeral, apical, and middle spots are made up of 3 humeral, 4 middle and 3 apical spots as shown in figure 290, plate XV, and 333, plate XVI. Thus he calls the markings which are most characteristic of the group the Marginal Component. He calls the median basal spot of the elytron the Basal Component (Bl) and the marking along the suture or anal border of the elytron the Sutural Component. He recognizes also such patterns as those sho\vu in figures 2-41, 243. 248, as Dis^pcrsion Component. He states that this analysis is for taxonomic purposes only and not based on ontogeny. He recognizes the most important tendencies toward joining of spots, in addition to the general plan outlined in G. Horn's four statements. The work of these men is here cited to show the fact that various generalizations have already been made sho^ving that the patterns eon- form to a general plan of spots or bands which have been similarly interpreted, though not exactly the same, by two authors with wide experience in the group. For the purposes of illustrating what may be determined in the / group in the way of general tendencies (p. 36) and the patterns of interrttpta, intcrrupta subsp. gahonica, flexuo.ia (PI. XII), tranque- ; iarka. and purpurea. And for a second illustration take the same species substituting .^cutfUari.'i for purpurea. First noting interrupta and e/aionka, (Figs. 156, 156 a and 165, and 165 a) one finds that the cross bands clearly recognized in Coleop- tera, especially Chrysomelidae, and Lepidoptera. and which appear in the tiger beetle group especially in the patterns associated with inter- rupta (PI. XII), and which appear in all the species in which ontogeny was studied, are present. In gahonica it appears that through iiidivid- 60 ILLINOIS BIOLOGICAL MONOGRAPHS [454 ual variatiou the characteristic joining to make the "middle band" is indicated. This occurrence of cross bands as noted and the variations of interrupta together with the light stripe in the region of joining of the cross band 4 with cross band 5.6 which occurs in the ontogeny of the patterns of scutellaris constitute the evidence for the line of development suggested. The second tendency to be noted is the shifting of the spots near the sutural or anal border of the elytron out of line with the cross band with which are properly associated. This is sho^vn in figures 156 and 156 a, plate XII, interrupta and in iigures 153 and 154 in flexuosa. The third tendency to be noted is the loss of the three small baso- sutural spots {Bl, C2.3, D4, Fig. 49, PL V). This usually takes place in a definite order if individual variation may be trusted as an indi- cator. At least these may have disappeared in some definite order leaving the typical pattern of tranquebari-ca as shown in the controls of the experiments (Figs 456a', i' and 457 a', &'). This type is shown in figures 31, 32, and 33, plate III, and the elements from which it is made are shown with others in figure 49, plate V. As further evidence of the longer persistence of C3.4 see figure 125, plate X, and figure 145, plate XI, which are late stages showing the persistence of this spot after the more anterior one has disappeared. The fourth tendency which may be noted is the tendency for the typical C. tranqueiarica pattern to shift as indicated in the patterns which result from experimental stimulation during ontogeny. This is shown in figures 456 a. h, 457 a, b, 458, 459, and 460 a, b, plate XXVIII. These modifications have already been noted on page 39 but may be recalled briefly as follows: the forward and backward extensions of the inner end of the humeral lunule (spot B2 di-ops out) disappear; the slight forward extension of the inner end of the middle band in the longitudinal stripe C{C5) drops out or loses identity. The with- drawal of the middle band from the elytral margin and reduction to conform Math that of C. purpurea (purpurea) (Fig. 537, PI. XXVIII) is the striking and probably the most important change best illustrated in 460 ff, b,. Similar modifications in all high temperature experiments with C. hirticollis (some with C. limbalis) serve to clinch the argument for response in definite directions. A fifth tendency is illustrated by C. purpurea as shown in the figures to the left in figure 537, plate XXVIII. The differences be- tween the purpiirea series and the tranquebarica series lies in the short humeral lunule of the former, which indicates a different tendency which perhaps constituted the original distinction between the patterns of tlie two series. 455] COLORS OF TIGER BEETLES—SHELFORD 61 Turniug to the scutellark series one notes that markings are re- duced by high temperature (Figs. 463 a, &,-. 464 a, 6, -, PI. XX). The original markings e\'ideutly included a middle band like purpurea (Fig. 512, PI. XXVIII). As e^•idenee for this note figure 490, plate XXVIII, which shows a reduced band present, and figure 115, plate IX, which shows one in ontogeny which does not persist in tlie adult at all in individuals from the central states. Stimulation of scutdlaris during ontogeny by high temperature merely reduces the markings concentriqly, withdrawing the middle band from the margin as well as from the centre. This is the type of modification which has led to immaculate forms in the south and southwest. Cold extended the same markings, but the results are not so strik- ing in general plan though perhaps equally general in application, as markings are lost in the same general order in many species if indi- vidual and geographic variation may be used as an indicator. First we have noted that purpurea is divided into two groups, one the steep- bank-inhabiting group and the other the level-ground-inhabitant. The latter (PI. XXV, left, and PI. XXVIII, Fig. 537) loses its markings in the manner suggested above, as indicated by the experimental results ■nith C. tranquebarka^ The outer end of the band being lost first. The other loses its martriugs as does C. smtellaris. Compare 486 to 490 with 506 to 510 and 522 to 525, plate XXVIII. which indicate the loss of markings of several species along similar lines, i.e., through retreat to the margin and then reduction of the marginal markings. Thus the response to high temperature represents a tendency present in many species. The large confluent markings of Manitoba specimens and of those which have been subjected to cold suggest that a second type of re- sponse may be in the form of a concentric extension of the unpigmented areas. It seems e\'ident that the mechanism in C. scutellaris may be thrown in either direction from the general average of the species. I have followed through a series of marking modifications and shown evidence for the tendencies indicated. It would be futile to present further discussions of a similar type regarding other species, as particidar weight is given to experimental results and such results are wanting in other species. Tlie reader by an inspection of the figures which are particularly numerous and selected for the purpose will note that in many groups one species begins in pattern modifica- tion where another leaves off. This fact was noted by G. Horn (1892). In many cases an exact knowledge of the geographic variation of the species is not available, but figures 435 to 437, plate XVIII, show a series which is supported geographically. C. curvata which occurs in Mexico 62 ILLLXOIS BIOLOGICAL MOSOGRAPHZ [456 is first ill the series, dorsalis saulcyi which occurs iii Texas next, and dormlis which occurs in New York and New England shows spreading of the white. This series is representative of one in which the patterns are of a specialized type, in which the media, trachea is reduced. For- ward curves in the humeral lunule are veVy rare ; one specimen of saulcyi in the collection of Mr. Gestroi in Genoa has this marking curved forward. The backward curvature occurs also in trifasciata peruviana but is rare. Figure 434, plate XVIII is probably this species. Much detailed study and collecting is necessary to show that the differences which enable one to arrange a group of pattez'ns in series really represent a series geographically or habitudinally separated, and the writer refrains from further discussion of such cases though others might be cited with little doubt as to their validity. The patterns in the illustration pages are arranged to show probable lines of modifica- tion. The large series of parallel trends shown in different groups leaves little doubt that the tendencies shown are highly probable. Another tendency quite common in the Cicindelas is the degener- ation of the media trachea. The shifting of the pattern in that region is one of the first modifications to take place if we may judge from the existing patterns and from individual variation. The complete break- ing up of the system of markings appears first in this part of the elytron. This degeneration of the old system of markings has pro- ceeded far in some species such as figure 16, nivea and figure 21, tenuipes. Here an almost entirely new system has grown up, but de- rived from the older one. These cases constitute our best evidence that these patterns are highly specialized. The morphological struc- tures with which the pattern is associated, are modified ; some of the important parts have degenerated. In considering these patterns and the modifications which take place the reader must not fail to note that there are physieological problems to be considered and physiological work to be done. The explanation for the occurrence of pigment in some parts of the body and not in others may be very simple. In course of experiments con- cerned with the production of abnormalities, it was found that the labrum which is not pigmented in the species used, develops pigment in the area of wounds. Specimens with abnormal elytra which appear to be due to injury or irritation nearly always have reduced patterns, but no cases in which the white markings are extended are recorded. Tlras it appears that the present adult areas of pigmentation and areas of ontogenetic and earlier pigmentation may be merely areas occupied by cells with a higher rate of metabolism. This in the normal elytron 457] COLORS OF TIGER BEETLES— SHELFORD 63 may be due to advantageous nutrition conditions arising from the mor- pliology of the wing, or to special characteristics of the cells themselves. BEARING OF THE COLOR PATTERN MECHANISM ON ORTHOGENESIS Orthogenesis is commonly understood as evolution in certain direc- tion as opposed to evolution due to the survival of certain kinds of variations out of a large fortuitous, series. The chief points in the original contention of Eimer, namely, that progress in species formation has been along definite lines, has been so generally admitted that the remaining matters are concerned with such questions as: Hoav definite have the directions of modification been? What are the causes of cer- tain-directions of modification being developed to the exclusion of others? Are the causes external or internal? Whitman has empha- sized the internal causes, which is the tendency of all who come at the problem from the point of view of embrj'ology, cytolog.y, and modern genetics. The mystical nature of the question of the origin of a com- 'i \ plex organism from a single cell, transmitted through the egg and the ' / sperm of tlie entire series of details which are inherited, have fascinated men and led to the general acceptance of theories which' involve the insulation of the bearers of herditary characters from the environment. The evidence at hand does not justify any detailed discussion of this problem but I will turn to the few things which appear to apply to the tiger beetle group. The effects of high temperature on tranqucbarica produce varia- tions in the direction of shortening the longitudinal portion of the middle band and throwing this marking into an oblique position. This is also one of the general tendencies in a large group of tiger beetles. In tranqufharica it occurs as a response to stimulation, and in its races of luiknown stability in regions in which high soil temperatures may be expected. It occurs in nearly half the species of the group of tiger beetles as a regular, probably hereditary character. Tlie condition of the middle band seems to be due to a mechanism of response or modi- fication, which is the same in these responses to stimuli and in the regular heredity trends. The problems of herdity then appear to be the same as the problems of development and modification of this elytral character of Cicindela. Perhaps the weakest point in the en- tire method of study and reasoning of tliose interested in problems of heredity is tlie apparent practical assumption that laws of heredity are not the same as laws governing characters, in particular organs, and as laws of response. Tlie evidence presented tends to show that these laws are one and the same and are dependent upon a mechanism present in the elyti*a of many species of Cicindela. If tliis is what is 64 ILLINOIS BIOLOGICAL MONOGRAPHS [458 meant by ortliogensis this group illustrates the orthogenetie priuciijle. The illustration above is concerned, however, with oul.y one of sev- eral kinds of tendencies which appear in the group. Still another Ijrinciple is suggested by the experiments. If extension of the unpig- mented areas is indicated by the experiments with cold conditions dur- ing ontogeny, which would be supported by geographic variation in many species, one is forced to the conclusion that different kinds of stimuli acting on the pattern mechanism produce different responses. One type of response is the extension of the unpigmented ai"eas. From an inspection of the figures it appears that this may take place on the basis of a pattern in any stage of reduction. As a rule it occurs in cor- relation with some marked change in the basal structures of the elytron at least when the extensions violate the original plan of the pattern. The mechanisms of pattern heredity and pattern development possess the capacity both to respond to stimuli by changes in form and by the extension of the unpigmented areas. This extension of the unpigmented areas may take place in almost any form of pattern shown in the entire series and may be concentric or in part linear. This is shown in plates XII to XVIli and XXXVIII. The concentric extension at least would seem to constitute a sort of reverse principle to that illustrated by the changes in form resulting from my experimental conditions, such as high temperature. In dealing with definite directions of response which may be termed orthogenetie if desired, one must recognize pro- gressive modification on the basis of a mechanism which may move in any one of two or three or more directions under the stress of external stimuli. Some evidence for a progressive series of modifications in the same direction running through a series of species in the tiger beetles is afforded by the experimental results. In general tlie pattern of C. hirticollis is more angular and as a whole conforms to the original groimd plan better than tliat of C tranquebarica. The modification of the patterns (middle band) of C. hirticollis is in a direction toward that of C. tranqueharica, but is not carried so far as are the modified patterns of C. tranqueharica. C. limhalis is usually, in the less modi- fied forms of middle band, about as far from the original angular type as are the more modified forms of C. tranquebarica. Stimulation of the mechanism of the middle band in limbalis at this stage usually throws the band still further toward that of splendida or typical pur- purea (Fig. .537, PI. XXVIII). Since the middle bands of the three species differ normall.y only in the extent to which such oblique shift- ing occurs, and each differs from the original plan to a greater degree than the other, the peculiar character of the direction taken must re- sult from a similarity of mechanism in tlie different species concerned. Abundant evidence for stages in such shifting as fixed hereditary 459] COLORS OF TIGER BEETLES—SHELFORD 65 characters is found iu many patterns illustrated in the plates, particu- larly plate XXXVIII. The series of three species thus show the same tendency, with respect to the middle band. This must be due to the existence of the same mechanism for heredity and response. The next step to important discovery probably lies in the direction of further analysis of the mechanism by experimental means, which may include surgical and mechanical experiments on the developing wing covers, and analysis by such methods are commonly used by the breeder. BEARIXG OF THE PATTERN MECHANISM ON THE BIOGENETIC LAW The data accumulated in connection with this study shows certain principles'''coficerned with the application of the biogenic law. First the general plan of the pattern seems to be common to all insects. The ancestry of the insect group is too obscure to justify the assumption that any original ancestor possessed a wing with nineteen spots such as arei shown in the elytron of Cicindela, or that such an ancestor possessed longitudinal stripes or cross bands. The evidence seems to indicate that the tiger beetle group sliows a type of pattern mechan- ism described at length in the preceding pages; that this pattern mechanism is plastic at least in the more generalized species; that from this plastic mechanism certain definite lines of modification have been somewhat fixed and limited. So long as the ontogeujie features are concerned with the general mechanism one is not justified in calling the appearance of certain spots recapitulations. They may full.v as well be areas which are less favorably nourished or which are made up of cells with lower rates of metabolism (see p. 31). Either of these physiological conditions may be due to mechanical necessities in devel- opment in all insects primitive and specialized, and if so, why call them recapitulations 1 Such evidences of recapitulation as do occur are found in the re- currence of markings in development which represent those occurring in related species or varieties. Thus, as I have noted, a curved middle band occurs in the ontogeny of some specimens of C. Iccontri and du- plicates a late stage in the loss of this marking as sho^ni in figure 115, plate IX. Here a curved and degenerate form of this marking occurs temporarily during ontogeny and may perhaps be regarded as recapitu- lation. The application of the biogentic law must generally be fol- lowed with great caution in dealing with insect patterns and no doubt with many other phenomena. 66 ILLINOIS BIOLOGICAL MONOGRAPHS [-WO SUMMARY OP CONCLUSIONS 1. The color patterns of the tiger beetles are related to elytral structures but not casually; longitudinal stripes in which pigment usu- ally occurs lie in the area of the chief tracheal trunks ; there are seven cross bands in which pigment does not develop, the second and third and fifth and sixth of these are often joined to make one of each pair. 2. Pigment usually occurs about the bases of hairs which usually lie in the lines of the tracheae. 3. In ontogeny the elytra show a spotted condition corresponding to the system of cross bauds and longitudinal stripes. The longitudinal stripes are usually more pronounced. 4. The characteristic markings of the group are composed of spots or elements joined in the longitudinal light stripe areas and areas of cross bands with the loss of various spots or elements which occur in ontogeny; joinings are sometimes oblique and when so markings are sometimes parallel with curved end of the elytron. 5. Certain particular types of markings made up of a few elements joined in a particular way characterize the majority of species of the group. 6. These markings as derived from the cross and longitudinal bands are angular; reduction of angles, straightening and turning into oblique positions parallel with the end of the elytron characterize modi- fications of markings. The response to stimuli (high temperature) is in the same direction. 7. Response to other stimuli appeal's to be in tlie direction of concentric extension of the markings. 8. The color patterns and structure to which they are related constitute a mechanism, the directions of movement of which are lim- ited, i.e., easier in some directions than others; the color pattern plans bi-eak when the related structures do; hereditary changes and fluctua- tions due to stimvilation during ontogeny are in the same direction; laws governing the mechanism are the same throughout. 9. These laws when applied to hereditary changes are apparently what is sometimes termed orthogensis. 10. It is not correct to assume that all manifestations of the ■wing mechanism which appear during ontogeny follow the biogenetic law. 461] COLORS OF TIGER BEETLES— SHELFORD 1. The brilliant colors of the group are due to thin surface films of material having properties of metals. 2. Changes in color during ontogeny are from green and blue toward red or brown, except in C. lepida in which it is from yellow (gold) to green; purples appear to stand apart from greenish blues and do not change diiring ontogeny or if so only slightty. 3. During ontogeny some species pass through stages correspond- ing to geographic races, but the biogentic law is of doubtful applica- tion, though green stages in ontogeny possess the same amount of pig- ment as green races and the reds and brown which come later are as- sociated with more pigment but not causally. GEOGEiVPHY 1. The center of distribution of the group is about the Indian Ocean. 2. Geographic races and geographic distribution Js^uot correlated .A.^-^ ■ with any observed climatic or- meteorological conditions unless it be rainfall and in this case the correlation is not complete. This lack of correlation is believed to be due to a lack of records of soil conditions. 3. Experimental modifications nearly duplicate certain geographic races of the species concerned ; these races occur in localities where conditions are probably similar to the experimental condition. 4. In the species studied in detail the more brilliant colors are in warm arid localities, reduced marking in warm localities, and ex- tended marking in cooler localities. 68 ILLINOIS BIOLOGICAL MONOGRAPHS [462 BIBLIOGRAPHY Adams, C. C. 1902. Southeastern United States as a Center of Geographic Distribution of Flora and Fauna. Biol. Bull,, 3:115-131. Bates, H. W. 1884. Coleoptera, Cicindelidae. Biologia Centrali-Americana, Insecta, i, pt. I, 1-18. Braun, Annette F. 1914. Evolution of the Color Pattern in the Microlepidopterous Genus Lithocolletis. Jour. Acad. Nat. Sci. Phila., (2)16:105-168. Bachmetjew, p. igo2. Experimentelle Entomologisches Studien. i :Leipzig. 1907. Ibid. 2;Sophia. CoMSTocK, J. H., and Needham, J. G. 1898. The Wings of Insects. Amer. Nat., 32:43, 81, 231, 335, 413, 561, 769, and 903. Child, C. M. 1915. Senescence and Rejuvenescence. Chicago. Criddle, N. 1907. Habits of Some Manitoba- Tiger Beetles. Can. Ent., 39:105-114. 1910. Ibid., 42 :9-l6. Heylaerts, F. J. M., Jr. 1870. (On Cicindcia ca)iif'cstris). Tijdschrift voor Entomologie, 13:178- 179. EiMER, G. H. T. 1898. On Orthogenesis and the Importance of Natural Selection in Species Formation. Religion of Science Library, Chicago. Enteman, W. 1904. Coloration in Polistes. Carnegie Inst. Wash., Publ. 19. FOLSOM, J. W. igo6. Entomology with Special Reference to its Economic and Biological Aspects. Philadelphia. Gortner, R. a. 191 1. Studies in Melanin. Amer. Nat., 40:743-755. Horn, George 1892. Variation in Cicindelidae. Ent. News, 3 :25-28. Horn, W. 1904. Cicindelidae of Ceylon. Spolia Zeylanica, 2:1-16. 1905. Systematischer Index der Cicindeliden. Deutsche Ent. Zeit., Supp. Feb. 463] COLORS OF TIGER BEETLES—SHELFORD 69 Horn, W. {continued) 1906. Beitrag zur Erkenntnis der Zeichnungs-Abanderungen bei Cicindeli- den. Deutsche Ent. Zeit, 173-174. 1908. Genera Insectorum. Carabidae ; Cicindelinae, fasc. S2A. 1 910. Ibid., fasc. 82B. 1915. Ibid., fasc, 82C. Horn, W., and Rosche, H. 1891. Monographie der palaarktischen Cicindeliden. Berlin (published pri- vately). Lexg, C. W. 1902. Revision of the Cicindelidac of Boreal America. Trans. .Am. Ent. Soc, 28:93-186. 1912. The Geographic Distribution of Cicindelidae in Eastern Xorth .Amer- ica. Jour. X. Y. Ent. Soc, 20:1-17. Linden, ^I. von 1902. Le dessin des ailes des Lepidopteres, recherches sur son evolution dans I'ontogenese et la phylogenese des especes son origine et sa valeur systematique. Ann. sci. nat. zool., (8)14:1-196. M.WER, A. G. 1896. The Development of the Wings and Their Pigment in Butterflies and Cloths. Bull. Mus. Comp. Zool., 29:209-235. 1897. On the Colors and Color Patterns of Moths and Butterflies. Proc. Best. Soc. Nat. Hist, 27:243-330. MiCHELSON, .-\. A. 191 1. On Metallic Colouring in Birds and Insects. Phil. Mag., 554-567. P.XCK.^RD, -A. igoo. Textbook of Entomology, New York. Riddle, O. 1909. Our Knovv'ledge of Melanin Color Formation and Its Bearing on the Mendelian Description of Heredity. Biol. Bull., 16:316-351. Shelford, V. E. 1906. Horn's Systematischer Index der Cicindeliden. Jour. X. Y. Ent. Soc, 14 0-9. 1907. Preliminary Note on the Distribution of the Tiger Beetles (Cicin- dela) and Its Relation to Plant Succession. Biol. Bull., 14:9-14. 1908. Life Histories and Larval Habits of the Tiger Beetles. Linn. Soc. (London) Jour. Zool., 30:157-184. igil. Physiological Animal Geography. Jour. Morph. (Whitman Volume), 22:551-617. 1912. Ecological Succession, IV. Vegetation and the Control of Land Com- munities. Biol. Bull., 23:59-99. 1912a. Ecological Succession, V. .Aspects of Physiological Classification. Biol. Bull., 23:331-370. 1913. Noteworthy Variations of the Elytral Tracheatinn of Cicindcla. Ent. News. 24:124-125. 1913a. The Life-History of a Bee-Fly (Sl'ogostyltim anale. Say) Parasite of the Larva of a Tiger-beetle (Cicindcla scutcltaris lecontei, Hald.) .Amer. Ent. Soc, 6:213-225. 70 ILLINOIS BIOLOGICAL MONOGRAPHS [464 Shelford, V. E. (continued) 1913b. Animal Communities in Temperate America as Illustrated in the Chicago Region. A Study in Animal Ecology. 380 pp. Chicago. 1914. Abnormalities and Regeneration in Cicindela. Amer. Ent. Soc, 8:291-295. 1915. Elytral Tracheation of the Tiger Beetles (Cicindelidae'). Tr. Amer. Micr. Soc, 34:241-254- Tower, W. L. 1903. Origin and Development of the Wings of Coleoptera. Zool. Jahrb., Anat., 17:517-572. 1903a. Colors and Color Pattern of Coleoptera. Dec. Publ. Univ. Chicago, (i) 10:33-70- 1906. An Investigation of Evolution in Chrysomellid Beetles of the Genus Leptinotarsa. Carnegie Inst. Wash., Publ. 48. Whitman, C. O. 1904. The Problem of the Origin of Species. Cong, of Arts and Sci., Universal Exposition, St. Louis, 5 :4i-58. WiCKHAM, H. F. 1902. Habits of American Cicindelidae. Proc. Davenport Acad. Nat. Sci., 7 : 206 -228. 1904. The Influence of the Mutations of the Pleistocene Lakes upon the Present Distribution of Cicindela. Amer. Nat., 38 :643-654. 1906. The Races of Cicindela tranqKcbarica Herbst. Ent. News, 17:43-48. Zeleny, C. 1907. The Direction of Differentiation in Development. The Antennule of Mancasellus macroyrus. Arch. Entw. ]\Iech. Org., 23 :324-343. 465] COLORS OF TIGER BEETLES— SHELF ORD EXPLANATION OF PLATES Because of the diversity of material studied, the plates of this monograph have been made in different ways and for details and excep- tions to the general statements below it will be necessary to see the text. Plates I to IV show camera drawings made chiefly by mounting dry elj^ra in hot balsam containing little or none of the usual solvents. Plate V is a diagram. Plates VIII to XI were made from specimens killed in the best of fixing fluids and mounted according to approved methods. They represent different indi\'iduals chosen at different stages, but have been checked with individual histories. Plates VI, VII, and XII to XXVIII, in so far as they are concerned with eh-tra, are made up of free-hand drawings of elytra as seen from directly above the center of the curved side, i. e. to the left and above the specimen. The specimens represented are from various sources. All are dra^ni the same size though the specimens vary greatly. The drawings in plates XII to Xr\"III are about twice the natural size of an average species. The distribution data shown were supplied from various col- lections and printed lists. The colored plates which show color ontogeny were made chiefly from the same living individual. 72 ILLIXOIS BIOLOGICAL MOXOCRAPHS [466 PLATE I Explanation of Plate Figure i. Cross section of the adult elytron of C. lepida, showing the re- lation of lack of pigment to interlamellar spaces. The portion at the right is through a pigmented area and that at the left through an unpigmented area. PCU, primary cuticula, unpigmented ; PCP, primary cuticula, pigmented ; SC, secondary cuticula. The portion under the unpigmented areas is divided into layers sepa- rated by air-tilled spaces above which small canals project into the layer above ; under the pigmented part the cuticula is in clear layers with no spaces between. The air spaces in the cuticula imder the unpigmented portion are probably the cause of the appearance resembling white pigment in the unpigmented areas. Figures 2-9. Showing the relation of the markings and tracheae in Cicindela. The tracheae present are from left to right casta. (Co) (see Fig. 21), the subcosta (S), the radius (R), the media (M), and the cubitus (Cu). The anal cannot ordi- narily be demonstrated in dried elytra. The drawings were made with a camera lucida. The figures indicate a number of unpigmented areas which are in the form of cross bands which may be broken by the pigment lying in the lines of the tracheae ; the letters A, B, and C indicate the unpigmented stripes to fall be- tween the tracheae. Fig. 2. C. rcgalis Dej. (Africa) : 3, intcrrupta Fabr. (Africa) ; 4, intcr- rupta Fabr. (Africa) ; 5, dongalensis Klg. (N. Africa) ; 6, vigintiguttaia Herbst. (India); 7, compressicornis Boh. (Africa); 8, comprcssicornis Boh. (Africa); 9, discrepans Walk. (Ceylon). ILLINOIS BIOLOGICAL MONOGRAPHS PCU SHELFORD COLORS OF TIGER BEETLES PLATE I 74 ILLINOIS BIOLOGICAL MONOGRAPHS [468 PLATE II Figures 10-21. Showing, the close correlation between the distribution of tracheae and dark pigment in the more specialized patterns of Cicindela. Figures 10 to 22 show close conformation of color patterns and tracheae. » Explanation of Plate Fig. 10. C. striolata Illig. (India) ; 11, cincta Oliv. (Africa) ; 12, anchoralis Chvr. (S. China) ; 13. quadrilineata Fabr. (India) ; 14, capensis Linn. — the costa and subcosta were probably present but could not be demonstrated; 15, capensis Linn. sub. sp. chrysographa Dej. (S. Africa) ; 16, nivea Kirb. aber conspersa Dej., showing reduction of the media (S. America) ; 17, painphila Lee. (S. U. S.) ; 18, lugubris Dej. (Africa); 19, gabbi G. Horn (S. W. U. S.) ; 20, dorsalis Say (Coast of U. S. A.) ; 21, tcnuipes Dej. (India), Co, costa, S, subcosta, R, radius. M, media, Cu, cubitus. ILLISOIS BIOLOGICAL MONOGRAPHS VOLUME 3 SHELFORD COLORS OF TIGER BEETLES PLATE II 76 ILLINOIS BIOLOGICAL MONOGRAPHS [470 PLATE III Figures 22-33. Showing the transverse, longitudinal, and oblique bands in Ctenostomidae Collyridae; Cicindelidae (Dromicini and Odontochilini) and va- riations in the markings of C. tranquebarica, a species with typical patterns and variations. Explanation of Plate Fig. 22. C. longipes Fabr. (Malay Arch.) ; 23, imperfecta Lee. (S. W. U. S.) ; 24, luteolineata Chvr. (Mexico) ; 25, lemniscata Lee. (S. W. U. S.) ; 26, Ctenostoma obliguatum Chd. (South America), showing the central transverse band and distal spot; 27, Ctenostoma unifasciatum Dej. (S. America) ; 28, Collyris celebensis Chd. (Malay Arch.), showing three lighter cross bands; 29, Heptodonta analis Fabr. (India), showing spots representing two cross bands; 30, Dromica coarctata Dej. (S. Africa), showing longitudinal stripes and heavier pigment in the lines of the tracheae; 31-33, showing patterns of C. tranquebarica Herbst (N. A.), typical pattern (31) and extended pattern with extensions between the tracheae (32), and a reduced pattern (33) with the middle marking broken in the line of the trachea. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 4- 27 p ^1 COLORS OF TIGER BEETLES PL\'IK III ILLINOIS BIOLOGICAL MONOGRAPHS [472 PLATE IV Figures 34-38. Showing the relation of the tracheae to pigmentation of the elytra in Carabidae and Dytiscidae. Explanation of Plate Fig. 34. Omophron sp. (N. A.), showing suggestions of transverse bands num- bered to correspond with figures 66 and 67 and a tendency for white markings between the bands to lie between the tracheae ; 35, Benibidium versicolor Lee. (Illinois), showing the unpigmented areas in the lines with the tracheae; 36, unknown carabid (Amazon), showing the pigmented areas in the lines of the tracheae; 37,Nebia coiiiplanata Linn. (Europe), showing a tendency to lines over the tracheae and between them; 38, Hydacticus stagnalis Fabr. (Illinois), show- ing double lines ; 39, Laccophilus ntaculosus Say, showing the transverse bands and suggestion of double unpigmented lines with the tracheae in the pigmented areas (see Fig. 18) ; 40, Agabits tacniolatus Harr. (Illinois), showing trachea in the unpigmented areas with a suggestion of double lines ; 41, Hydroporus undn- latiis Say (Illinois), showing the cross bands — a suggestion of all those commonly present in Cicindcla. ILLINOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 COLORS OF TIGER BEETLES PLATE IV ILLIXOIS BIOLOGICAL MONOGRAPHS [474 PLATE V Figures 42-49. Showing an analysis of the color patterns of Cicindela. Explanation of Plate Fig, 42. Showing the full number of longitudinal stripes represented in the group — compare with figures 169, 169a, and 169^ Uctragraiiuna Boisd.) ; 43, show- ing the three longitudinal stripes nearly always represented — compare with 52 (C. tetragramina, variation) and 54, desgodinsi Fair (Tibet); 44, showing the splitting of the stripes as suggested in 53, lugubris Dej. (Africa) ; 45, showing the full number of cross bands numbered / to 7; 46, showing the commonest cross bands illustrated in 58 (rcgalis Dej. Africa) ; 47, showing a second com- mon type illustrated by 75, in which none of them reach clear across : 48, show- ing all the possible spots that can occur from a combination of the longitudinal stripes and cross band shown in figures 42 to 47 ; 49, showing the spots which are most commonly present or joined to form characteristic patterns in the group. A and a are usually fused on account of the crowding together of the traclieae. The cross bands are never all represented entirely across the elytron, but by dots as in 62, C. vigintiguttata Herbst (India). The fusion of Ai, As and Bs gives the characteristic humeral lunule of students of the group, the hook frequently present is made by joining it with B^. The fusion of Ci and Cs and of C3 and C4 gives the characteristic markings shown in the line C. of many old world species. The union of A4, B4 and 5.5 gives the characteristic middle band of the group. As is of rare occurrence (see Fig. 198). .J5 is commonly present as a spot, also A6, B6 and C6 are less common in occurrence (see Figs. 6 and 7, PI. I). ILLIXOIS BIOLOGICAL MONOGRAPHS VOLUME 3 COLORS OK TIGER BEETLES 82 ILLINOIS BIOLOGICAL MONOGRAPHS [476 PLATE VI Figures 50-77. Showing selected Cicindelid patterns with lines to show the correspondence of all the chief types of pattern to the plan shown in Plate V. Explanation of Plate Figs, so, 51, 52, C. tctragramma Boisd. (Australia) ; 53, lugubrus Dej. (Af- rica) ; 54, desgodinsi Fair (Tibet) ; 55, interruptofasciata Schm. (Siam) ; 56, inuata sub. sp. laticornis Horn (Africa) ; 57, coinpressicornis Boh (Africa) ; 58, rcgalis Dej. (Africa) ; 59, atkinsoni Gestro (Australia) ; 60, regina Kolbe (Africa) ; 61 melaleuca Dej. (S. A.) ; 62, vigintignttata Herbst (India) ; 63, notata Boh (Af- rica) ; 64, gcrstacckcri Horn (Africa) ; 65, Euryoda adonis subsp. rufosquata Bell (Madagascar), Boh; 66, damcnsis, (Siam); 67, Odontochila singularis Fit. (S. A.) ; 68, Peridexia hilaris, Fajrm. (Madagascar) ; 69, flavosignata, Cast, (Af- rica) ; 70, crespigjiyi Bates (Borneo) ; 71, anchoralis Schm. (China) ; ^2, copulata Schm. (India) ; 73, interriipta subsp. gahonica Bat. (Africa) ; 75, aphrodisia Baudi (Cypris) ; 76, aurtdenta Fabr. (India); 77, 6 punctata, Fabr. (India). IIJJXOIS BIOLOGICAL MOSOCRAPHS VOLUME 3 SHELFORD 73" ' ' 74" ' ' 75' ' ■ 76' COLORS OF TIGER BEETLES 77' PLATE VI ILLIXOIS BIOLOGICAL MONOGRAPHS [478 PLATE VII Figures 78-98. Showing some of the chief lines of union of markings not indicated on the preceding chart. Explanation of Plate Fig. 78, showing the spots which enter into the patterns with some of the characteristic unions indicated — the stippled areas refer to figures 79 and 80; the narrow white lines to 81, 82, and 83; the dotted lines to 88 and 89, and 90 and 91; 79, apiata dausseni Putz (S. A.) ; 80, striolata subsp. trisignata Chd. (India) ; 81, fatidica Guer (Africa) ; 82, (Prodotcs) miinula Per. (Africa) ; 83, viridis Raffr. (Africa); 84, pdetieri (N. Africa). Figs. 85-87. Showing the oblique shifting of the cross markings ; 85, regalis Dej. (Africa) ; 86, andriana All (Africa) ; 87, maheva Kunck (Africa) ; 88, cey- lonensis Horn (India) ; 89, oscari Horn (Africa) ; 90, kolbci Horn (Africa) ; 91, princeps ducalis Horn (India) ; 92, longipes Fabr. (Malay Arch.) ; 93, albicans Chd. (Australia) ; 94, nitida Wdm. (India) ; 95, trisignata Dej. (Europe) ; 96, nitidula Dej. (Africa) ; 97, gahhi S. Horn (S. W. U. S. A.) ; 98, Icuconoe, Bates (Mexico). Figures 99-100. Showing the color areas of the larvae for comparison with figures 101 to 105. Ventral side of the abdominal segment of a larva of C. tran- qucbarica. The areas are lettered as in figure loi ; 100, showing the color centers of the dorsal side of a larva. Compare with figure loi. The area with the spira- cles is the pleuron. ILLINOIS BIOLOGICAL MONOGRAPHS VOLUME 3 SHELFORD 99 100 COLORS OF TIGFR BEETLES PLATI-. VI I 86 ILLISOIS BIOLOGICAL MONOGRAPHS [480 PLATE VIII Figures ioi-iio. Showing the development of pigment in the legs and body of C. tranguebarica, Herbst. Explanation of Plate Figures loi-iio. Showing the pigment beginning at the posterior end of the body and moving forward except the trochanters which are pigmented at emer- gence; loi, 3 to 6 hours after emergence; 102, 8 to 12 hours; 103, 12 to 15 hours; 104, 24 to 36 hours. A, anterior band of pigment on the segment; /, the posterior band of the segment; aa, the large central anterior area — compare with figure 105. Figs. 105-108, showing the development of pigment in the dorsal side of the abdomen; 105, at emergence, showing the large dorsal spots beginning of tlie posterior segments ; 105a, after 3 to 6 hours, showing the fusion of the spots toward the center ; 106, 8 to 10 hours after emergence, showing the nearly com- plete abdominal pigment, the beginning of the pigmentation of the thorax, and the lines on the head; 107, showing the increase in the head and thoracic regions at 12 to 15 hours after emergence; 108, showing the dorsal side of the head and thorax after 24 to 36 hours; 109 o to d, the antenna; 3 hours after emergence; h, 6 hours after emergence; c at 8 to 10 hours after emergence; d, 11 to 15 hours after emergence; c, 24 hours after emergence; no, a, showing the hind leg three days after emergence; b. at emergence; c, after 6 to 8 hours; d, 12 hours after emergence. ILLISOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 SHELFORD COLORS OF TIGER BEETLES PLATE VIII 88 ILLINOIS BIOLOGICAL MONOGRAPHS [482 PLATE IX Figures 111-122. Showing stages in the development of pigment in the elytron of C. repanda Dej. and C. scutellaris aber lecontei Hald. Explanation of Plate Fig. Ill, 4 to S hours after emergence, showing the longitudinal lighter areas corresponding to A, B, C of the preceding figures; 112, after 12 to 15 hours, showing the stripes A, B, C broken into cross bands, 3 and 4 being clearly indi- cated in the stripe C; 114-118, showing stages in the development of the elytral pigment in C. scutellaris lecontei Hald; 114, after 4 to 5 hours; 115, after 12 hours, showing particularly a well indicated cross band not appearing in the adult; 116, after 15 hours, showing well marked longitudinal bands broken in spots; 117, after 36 hours with similar marking indicated; 118, after 36 hours, similar to 117; 119, the hind wing at emergence; 120, after 12 hours; 121, after ^6 hours; 122, adult. ILLINOIS BIOLOGICAL MONOGRAPHS VOLUME 3 COLORS OF TIGER BTiETLES PLATE IX ILLIXOIS BIOLOGICAL MOXOCRAl'IIS [484 PLATE X Figures 123-134. Showing stages in the development of the pigment of the elytra of C. purjmrea limbalis Klg. (123-130), and in C. tranguebarica Herbst (131-134). The wing areas are indicated bj' letters and numbers as in the pre- ceding figures. Explanation of Plate Fig. 123. Three hours after emergence, showing the lighter areas between the tracheae ; 124, after 8 hours — suggestion of both longitudinal stripes transverse bands; 125, showing a similar condition after 10 hours; 126, similar conditions at the end of 12 to 15 hours; 127, a similar suggestion of markings at 30 hours; 128, well defined markings at 36 hours ; 129. striking longitudinal stripes at 36 hours ; 130, heavier pigmentation in the lines of the trachea in the adult; 131 to 134, showing a similar series for the development of pigment; in C. tranqucbarica Herbst; 131, 6 to 8 hours after emergence; 132, 10 hours after emergence; 133, 12 hours after emergence; 134, 24 to 36 hours after emergence. lUJXOlS BIOLOGICAL .]JO.\OGRAPIfS 130 92 ILLIXOIS BIOLOGICAL MONOGRAPHS [486 PLATE XI Figures 135-146. Showing the ontogeny of pigmentation in C. punctulata Oliv., C. sexguttata Fabr., Tetracha Carolina Linn., C. hirticollis Say, and C. is guttata Dej. Explanation of Plate Fig. 13s, C. punctulata Oliv. at the end of 6 hours after emergence ; 136, after 12 to 13 hours; 137, after 36 hours; 138, sexguttata Fabr. after 24 hours; 139, Tetracha Carolina Linn, at the end of 9 hours after emergence ; 140, the adult elytron; 141-145, showing stages in the development of pigment in hirticol- lis Say; 141, 4 hours after emergence; 142, after 6 to 10 hours; 144, after 12 hours; 145, after 16 to 24 hours; 146, I2 guttata, after 12 to 18 hours. ILLISOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 m.- 135 136 138 f \ — ^ — — 2 r \ 4 ~5— ~6— \ 142 COLORS OF TIGER BEETLES ILLINOIS BIOLOGICAL MONOGRAPHS PLATE XII Figures 147-187. Showing patterns made up cf longitudinal and transverse bands variously broken and contrived. Follow the arrows in tracing out the dif- ferent directions of modification. For meaning of letters see page 9. EXPLANATIOX OF PlATE Figs. 147-149SJ showing tHe typical transverse wide transverse band type of pattern and modifications; 147, inahcva Kunck. (Madagascar); 148, andriana All. (Madagascar) ; 149, rcgalis Dej. (Africa) ; 149 a and b, the same. Figs. i50-i67a, showing internipta gahonica type of broken transverse bands and their modification. The patterns of gabonica 165 and 1650 are made up of transverse bands and broken in the lines of the tracheae with various lines of longitudinal and transverse union. Figs. 150-152, showing the unusual patterns belonging to this group; 150, singularis Chd. (N. E. Africa) ; 151, kollari Gistl. (S. Africa) ; 152, malaris Horn (S. A.); 153 and 154, fiexuosa Fabr. (Europe); 155, striatifrons Chd. (India); 156-1560, interrupta Fabr. (Africa); 157, montdroi Bat. (S. Africa); 158, brevi- coUis subsp. dathrata Dej. (Africa) ; 159-160, Candida Dej. (Africa) ; 161, blandi- ardi Fairm. (S. Africa) ; 162, pddkri Luc. (N. Africa) ; 163, vittigcra Dej. (India) ; 164, multiguttata Dej. (India) ; 165, interrupta Fabr. subsp. gabonica (Africa); 165a, interrupta Fabr. subsp. gabonica: 166, laetescripta Mtsch. (E. Asia) ; 167, lepida Dej. (Illinois). Figs. 168-1690, showing pattern with three longitudinal stripes; 168. quccns- landica Sloane (Australia) (After W. Horn) ; 169a, tetragramma Boisd. (Aus- tralia) ; 170-1700, muata subsp. laticornis Horn (Africa); 171, muata Horn (.\f- rica) ; 172, juno Horn (Africa) ; 173, 1730, viridis Raff (Africa) ; 174, giganica Raffr. (Africa) ; 175, prodotiformis Horn (Africa) ; 176, fatidica Guer. (Africa) ; 177, miscranda Horn (Africa) ; 178, gcrstaeckcri Horn (Africa) ; 179. rcgina Kolbe (Africa) ; 180-1800, b, mechowi Lued (Africa) ; i8i-i8ia, braszai Fit. (Af- rica) ; 182, ininula Per. (Africa) ; 183, quadrisiriata Horn (Africa) ; 184-1840, peliti Guer. (Africa); 174a, gigantea Raffr. (Africa); 185, junkeri Kolbe (Af- rica); 186, vittata Fabr. (Africa); 187-1870, congecnsis Fit. (Africa). ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 183 m 184. SHELFORD COLORS 01' TIGKR BEETLES It? 18/. PLATE XII 96 ILLINOIS BIOLOGICAL MONOGRAPHS [490 PLATE XIII Figures 188-240. Showing the domination of the central stripe (B), obliquity in the middle band reversed from the usual type. For meaning of letters see page 9. Follow the arrows in tracing out the different lines of modification. Figures 188 and iS8a show the reduced transverse bands — compare with 149. Explanation of Plate Fig. 188 and 1880, C. flavosiyiiata Cost. (Africa) ; 184. dk'cs Gory, after Gory (India) ; 190, aitrovittata Brll. (India) ; igi, ccyloneusis divcrsa Horn, after Horn (India) ; 192, ccyloneusis Horn, after Horn (India) ; 193, discrcpans Wak (In- dia) ; 194 and 194a, hanuandi Fit. (India) ; 195, 196, intcrruptofasciata Schm. (India); 197, assamcnsis Parry (India); 198-199, sianicnsis Fit. (India); 200, andrewesi vauritti Horn (India); 201, princcps ducalis Horn (India); 202, auro- fasciata Dej. (India) ; 203, anrofasciata Dej. (India) ; 204-2040, anrofasciata lepida Gory (India); 206, assamcnsis ?; 207, crcspignyi Bat. (Malay Islands) ; 208, kachovskyi Horn (.Africa) ; 209, oskari Horn (Africa) ; 2io-2loa, shivah Parry (India); 211-212 (After Schaum), calligramma Schm, (India); 213, aurofasciata Dej. (India) ; 214, hacinorrhoidalis Wdra. (India) ; 215, burnicisteri Fischer (Asia) ; 216. stenodora Schm. (Malay Arch.) ; 217, minuta Oliv. (India) ; 218, craspedota. Schm. (Borneo) ; 219, semperi Horn (India) ; 220, caUigramnm Schm. (India) : 221, Prothynia adonis rufosignata Brll. (Madagascar) ; 222, chi- ncnsis japonica Thnb, (Japan) ; 223, chincnsis DeG, (China) ; 224, duponti Dej, (India) ; 225, e.vima Vand. (Malay Arch.) ; 226, fcfriei Fit. (Japan) ; 227, didyma Dej. (Malay Arch.); 228, aurulcnta Fabr. (India); 229, iiotata Wdm. (India); 230, aurulcnta Fabr. (India) ; 231, punctata Fabr, (India) ; 232, Thcratcs white- hcadi Bates (Malay Arch,) ; 233, T. fruhstorferi ?Iorn (Tonkin) ; 234, T. spini- pennis Latr. and Dej. (Malay Arch.) ; 235, T. chaudoiri Schm, (Malay Arch,) ; 236, T. tiiaindroni Horn (Malacca); 237, T. crir.ys Bates (Malay Arch.); 238, Pcridoxia hilaris Fair. (Madagascar) ; 239, Pcridoxia ftdvipcs Dej. (Madagascar) ; 240, Pometon singnlaris Fit. (S, A.). i ILUXOIS BIOLOGICAL MOXOGRAPHS' J'OLUME 3 234 I "5 23t, 237^ 238 239 2*« SHELFORD COLORS OF TIGER BEETLES PLATE XIII 98 ILLINOIS BIOLOGICAL MOXOGRAPHS [492 PLATE XIV Figures 241-283. Showing patterns made up of numerous spots and stripes. Figures 241 to 243 and 248, 2480, and 248b should be compared with plate IV, figure 38. In comparing the figures follow the arrows. For meaning of letters see page 9. ExPL.'iN.^TION OF Pl.ME Figs. 241-241(7^ b, C. comprcssii'ornis, Beh. (Africa); 242, kolbci Horn (Af- rica); 243, deyrollei Guer. (Africa); 244, maino Gestro ( N. Guinea); 245. atkin- soiii Gestro (India) ; 246, feisthameli Guer. (Africa) ; 247 (after Guerin)-247a, 7iysa Guer. (Liberia) ; 248, 2480, 248&, lugubris Dej. (Africa) ; 249, deyrollei Guer. (Africa) ; 250, vittata Fabr., after Guerin (Africa) ; 251, 20 guttata Herbst (In- dia) ; 252, desgodinsi Fair. (Tibet) ; 253, latreillei Guer. (Kapaur) — the stippled spots are dark and represent areas in which spots usually occur ; 254-255, rasti- cana Per. (S. Africa); 256, notata Boh. (S. Africa); 257, latreillei Guer. (Ka- paur) ; 258-258(7, b, rasticana Per. (S. Africa) ; 259, rasticaita aber cgregia Per. (S. Africa) ; 261, bioncani subsp. licngnici Per. (S. Africa) ; 262-263, striolata 111. (Burmah) ; 264, striolata subsp. trisignata Chd. (Timor); 265, neut-.ianni Kolbe (Africa) ; 266, pudica Boh. (Zulu) ; 2660, Boh. (Transvaal) ; 267-268, escheri Dej. (S. Africa); 269, nwrginella Dej. (Africa); 270, striolata 111. (In- dia) ; 271, do subsp. trisignata Chd. (Timor) ; 272, luxcri Dej. (Africa) : 273, heros Fabr. (Malay Arch.) ; 274-274^, hcros Fabr. (Malay Arch.) ; 275-2770, montciroi Bat. (Africa) ; 278-279, strachani Hope (Africa) ; 280-2800, b. cques- tris Dej. (Madagascar) ; 281, nitidula Dej. (Africa) ; 282, nilotica Dej. (Africa) ; 283, albina Wdm. (India). //././ A O/.V BIOLOGIC A L M OX OGRAI'lIS fi I-Q/A'ME 3 SHELFORD COLORS OF TIGER BEETLES PLATE XIV 100 ILLISOIS BIOLOGICAL MONOGRAPHS [494 PLATE XV Figures 284-328. Showing the patterns of North American species belonging chiefly to the Mexican and C. argentata groups and having cross bands 5 and 6 both distinctly represented in the majority. For meaning of letters see page 9. Various combinations of spots which go to make up the oblique vitta of some of the species of the group are represented in figures 291, 296, 297, 311, 312, 313, 319, and 320; compare these with figures 23 and 24 and 78 to 82. Explanation of Plate Fig. 284, C. polita Lee. (Texas); 285, abdominalis Fabr. (Atlantic coast, U. S.) ; 286, rufiventris aber. cumatilis Lee. (Texas) ; 287, rufiveiitris Dej. (East- ern U. S.) ; 288, 16 punctata Klg. (N. Mex.) ; 289, carthagena subsp. hentsi G. Horn (Mass.); 290, 16 punctata Klg. (Mexico); 291-2910^ rufiventris aber. mcllyi Chd. (Mexico) ; 292, trifasciata Fabr. (S. A.) ; 293-2930, h, trifasciata subsp. sig- nwidca Lee. (S. U. S.) ; 294, carthagena Dej. (Mexico) ; 295, rufiventris subsp. 16 punctatd Klg. (Mexico); 296, rufiventris aber mellyi Chd. (Mexico); 297, niclaleuca Dej.(S. A.); 298, obsoleta Say (S. W. U. S.) ; 299, fera Chv. (Mex- ico) ; 300, pusilla subsp. cinctipennis Lee. (S. W. U. S.) ; 301, punctulata Oliv. (U. S. and Mex.) ; 302, argentata subsp. aureola Klg. (S. A.) ; 303-3030, argentata Fabr. (Brazil) ; 304, lunalonga Schm. (California) ; 305, celeripes Lee. (Central U. S.) ; 306, cursitans Lee. (Miss. Valley); 307, schaupii G. Horn (Texas); 308- 308a, nephelota Bat. (Mexico) ; 309-3090, chlorostricta subsp^ staudingeri Horn (S. A.) ; 310, argentata subsp. venustula Gory (Mexico) ; 311-3110, pusilla subsp. imperfecta Lee. (Pacific States) ; 312, luteolineata Chvr. (Mexico) ; 313, lem- niscata Lee. (Arizona) ; 314, debilis Bates, after Bates (Mexico) ; 315, favergeri Brll., after Andouin and Brulle (S. A.) ; 316, 3160, 317, roseiventris Chvr. (Mex- ico) ; 318, flavopunctata Chvr. (Mexico) ; 319, pu^silla subsp. imperfecta Lee. (Pa- cific States) ; 320, crarcri Thms. (Mexico) ; 321, marquardti Horn — the only Cicindelid without a middle band (Sao Paulo) ; 322, hoegei Bat. after Bates (Mexico); 323-323(7, h, sommeri Mann. (Mexico); 324, anulipes Horn (S. A.); 32s, flavopunctata Chvr. (U. S. and Mexico) ; 326, chrysippe Bates (Mexico) ; 328, severa Laf. (Gulf States and N. M.) ; 328, striga Lee. (Florida). ILLINOIS BIOLOGICAL MONOGRAPHS VOLUME 3 323 323^ 3234 SMELFORD COLORS OF TIGER BEETLES PLATE XV 102 ILLINOIS BIOLOGICAL MONOGRAPHS [496 PLATE XVI Figures 329-377. Showing the patterns of the principal Eurasian species ex- clusive of the flcxuosa and longipes-biramosa-liiiiosa (Oriental) groups, with a few representative patterns from the genera Megacephala, Distypsidera, Cteno- stoma, and Collyris. For meaning of letters see page 9. Figures 330-332 are related American, species. Figures 329-355 show the typical and characteristic patterns of the genus Cicindela in which the portion of the elytron nearest tlie scutellum is without spots, in which bands 3 and 3 are fused and 5 and 6 are separate, and the modifications of the same. , ExPL.\NATio.\ OF Plate Fig. 329, C.donegalensis Klg. (Africa) ; 330, hirticolUs Say (Illinois) ; 331, re- panda Dej. (Illinois) ; 332, is guttata Dej. (Illinois) ; 33^,, lunulata Fabr. (Europe) ; 334, aphrodisia Baudi (Cypris) ; 335, lacryiiwsa Dej. (Japan) : 336, 10 guttata Fabr. (Malay Arch.); 337, discreta Schm. (Malay Arch.); 338, nitlda Wdm. (India) ; 339, contorta Fisch. (Europe) ; 3^)0, trisignata Dej. (Europe) ; 341, litterifcra Chd. (Europe) ; 342, alboguttata Klg. (Arabia) ; 343, suinatrensJs Herbst (India); 344-34S, orientalis Dej. (Europe); 346, melancholka Fabr. (Europe and Africa) ; 347, ,? signata aber subsuturalis Souv. (Europe) ; 348. circiiindata Dej. Europe) ; 349, circumdata Dej. (Europe) ; 350, anguhita Fabr. (India) ; 351, sumatrensis Herbst. (Oriental Region) ; 352, despecata Horn, after Horn (Madagascar) ; 353, ancosisconeusis Harris (New York) ; 354, funerca subsp. opigrapha Dej. (New Guinea) ; 355, variolosa Blanch. (Salathy) ; 356, galathea Thiem. (Asia) ; 357, lyoni Vig., after Roske (Europe) ; 358, 359, 359(7, germanica Linn. (Europe) ; 360. atrata Pall. (Europe and Asia) ; 361, 3610, h, germanica subsp. ohliquefasciata Ad. (Europe) ; 362, lactcola Pall. (Asia) ; 363, gennnata subsp. potanini Dok., after W. Horn (Tibet) ; 364, purpurea subsp. liiubalis Klg. (Illinois) ; 365, campestris, showing an unusual light area — the stippled portions are dark areas with cuticula such as covers the lighter spots; 366; ismenia Gory — stippled areas as in 365; 367, inaura Linn. (Europe); 368-369, fischeri Adams (Europe) ; 370, Megacephala auslralasiae huineralis McL. (N. W. Australia) ; ,371, quadrisignata Dej. (N. Africa); 372, M. (Styphloderma) asperata Wat. (Africa) ; 373, Distypsidera flavipes McL. (Australia) ; 374, D. gruti Pasc. (Aus- tralia) ; 375, Nickerlea distypsidcroides Horn, after Horn (Australia) ; 376, Ctenostoiua inacuUcorne Chvr. (Mexico) ; 377, Collyris frushtoferi Horn (Ton- kin). If.I.'XOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 " 3*63 364 365^' 366°^/ HtGACEPHAU DISTVPSIDtRA^^<;^ 3r,; 368 ^^ 3C9 370 371 ^ 372 " 373^^ 374 375 "" 376 377 SHELFORD COLORS OF TIGER BEETLES PLATE XVI 104 ILLIXOIS BIOLOGICAL MONOGRAPHS [498 PLATE XVII Figures 37S-421. Showing the patterns of the characteristic groups of species belonging to the Oriental and Australian Regions. For meaning of letters see page 9. They are in general of a character such as is commonly designated as specialized but show some viniisual combinations of areas which tend to con- firm tlie general interpretation here presented. ExPL.^N.\TiON OF Plate Figs. .vS-37?a, b, C. araih^if'cs Schm. (Borneo); 379-j79a, copnlatci Schm. (India); 380, aiichoralis subsp. iJUiutatissiiiia Schm. (China); 381-382, ornata Fit. (India); 383, 384, 385, psammodroina Chvr. (China); 386, 387, 388, anchoralis subsp. punctatissima Schm. (China) ; 385, anchoralis Chvr. (China) ; 390-391, q'.iadrilineata subsp. renei Horn (India); 392-393, ypsilon Dej. (Australia); 394, rafflesia Chd. (Australia) ; 395-3950, albicans Chd. (Australia) ; 396, longipes Fabr. (Malay Islands) ; 397-397a, 4 lincata Fabr. (India) ; 398, 398(7, 4 lincata subsp. renei Horn (India) ; 399, singularis Chd. (Nubia) ; 400, longipes Fabr. (Malay Islands) ; 401-4010, wapleri Lee. (Louisiana) ; 402-402U, mucronata Jord. (Malay Islands); 403, pupilligcra Schm. (New Guinea); 404, limbaia Schm. (In- dia) ; 405, ivaindroni Horn (India) ; 406, biratiiosa Fabr. (India) ; 407, bellana Horn (India); 4o8,funerata subsp. barbata Horn (New Guinea); 409, tuberculata Falir. (.Australia); 410, tuberculata aber latccincta White (New Zealand); 411, harryi Whitei (New Zealand) ; 412-413, lO guttata Fabr. (New Guinea) ; 414, niastcrsi ?klcL. (Australia) ; 415, fcrcdayi Bates (New Zealand) ; 418, tuberculata Fabr. (New Zealand); 419, dunedcnsis aber zcal;cficldi Bates (New Zealand); 420, fcrcdayi Bates (New Zealand) ; 421, pcrhispida Brn. (New Zealand). ILLISOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 SMELFORD 7 ^-^ "" 419 4?0 --^ ■1^1 COLORS OF TIGER BEETLES PLATE XVII 106 ILLIXOIS BIOLOGICAL MONOGRAPHS [500 PLATE XVIII Figures 422-455. Showing the higlily specialized patterns of the South Amer- ican species belonging chieflj' to the cnfrasccns and argciitata groups of species. For meaning of letters see page 9. All the types have representatives in which pigment has almost entirely disappeared as a rule and there is a strong tendency for the area of the media trachea to degenerate along with the reduction of that treachea (see figures 16 and 20). EXPLAN.'^TION OF PlATE Figs. 422-4220, h, C. apiata Dej. (S. A.) ; 423, apiata aber clausseni Putz. (S. A.) ; 424, gormazi Reed. (Chili) ; 425, mixta Horn (Ecuador) ; 426, trifas- ciata Fabr. (S. A.) ; 427-4270, b, graphiptera Dej. (S. A.) ; 428-4280, after Chev- rolot 4286, patagonica subsp. cherubim Chvr. (S. A.) ; 430-4300, b, marginata Fabr. (Texas) ; 431-4310, b, nivea Kirby (S. A.) ; 430-4320, gabbi G. Horn (California) ; 433, trisignata Dej. (Asia) ; 434, unidentified species from Arica, Peru, in the Oxford University Museum ; 435, curvata Chvr. (Mexico) ; 436, dorsalis aber saulcyi Guer. (Texas) ; 437-4370, b, c, dorsalis Say (Mass.) ; 438, malaris Horn (Pebas, Amazons) ; 439-4390, ncvadica var. knausi Leng (Kansas') ; 440, ciipras- cciis Lee. (Illinois) ; 441 hamata Brll. (Mexico) ; 442, chloroccphala Chv. (Vera Cruz, Mex.) ; 443, leucohoe Bat. (Mexico) ; 444, macroncma Chd. (Mexico) ; 445. togata Laf. (Texas) ; 446, atiraria Klg. (S. A.) ; 447, boops (West Indies) ; 448, macronema Chd. (Mexico) ; 449, pamphila Lee. (Texas) ; 450, togata Lee. (Texas) ; 451, californica praetexta Lee. (Texas) ; 452, marginata Fabr. (Texas) ; 453n, ''. c, d, 'cvapleri Lee. (Louisiana). ILLIXOIS BIOLOGICAL MOXOGRAPHS OLUME 3 449^^ •fbo 451 452 4<3j 453/, «b3<: 45Jj 454/ 454r SHHLF0RI3 COLORS OF TIGER BEETLES PLATE XVIII 108 ILLINOIS BIOLOGICAL MONOGRAPHS [502 PLATE XIX Explanation of Plate Figure 455. Showing the equipment used in the experiments on modification of tiger beetle color and color patterns. The experiments were conducted in four chambers ; two, A and B, which were of galvanized iron, rested with their bottoms in a concrete tank of running water. Trey were wrapped with cheese cloth and sprayed with jets of water on two sides which kept the mean temperature at 21° C. throughout the summer. The other two, C and D, were heated from above with electric lights, blackened in C and separated from the main chamber by a copper jacket in D. These were were heated to a point 10° C. above the temperature of the greenhouse except during the middle of the day. The switch shut off the heat at about 35° C. air temperature and the sun continued to heat the chamber so that the maximum soil temperature sometimes reached 40° C. or more by midafternoon. Air was drawn through the tanks by filter pumps and, in the case of the control tanks, through sulfuric acid for the dry one and water for the moist one, but this intake was not maintained for the high temperature tanks at all times because of mechanical difficulties. The moisture in the moist chambers was maintained by frequent additions of water to the soil, while in the dry chambers as little water as possible was added. H , sulphuric acid bottles ; S, mercury switch ; T, thermostat ; B, batteries ; W, water bottles ; SP, spray nozzles ; C, cloth cover. ILLISOIS BIOLOGICAL MOSOCRAPHS rOLCME 3 SHKLFORD COLORS OF TIGER BEETLES no ILLIXOIS BIOLOGICAL MOXOCRAPHS [504 PLATE XX Figures 456-465. Showing the color patterns of specimens of C. tranqiic- barica Herbst., C. piirpurca liiiibalis Klg., and C. sci:lcllaris lecontci Hald. sub- jected to high temperature under moist and dry conditions and placed in an ice box during their prepupal and pupal life. With them are shown controls which were kept at normal temperatures or lower and designated with letters a , b' , etc., and a few collected from the normal habitat from the same generation, desig- nated :c''. ExPL.\N.\TION OF Pl.ATE Fig. 456(J-(7, the elytra of seven specimens of C. tranquebarica which passed the late larval, prepupal, and pupal stages in a warm moist chamber ; mean tem- perature of the soil, 37° C. ; maximum for the warmest week, 40° C. ; control, 456a', b' , 'ii/, at 21° (Experiment 56) ; 457a-b, the elytra of two specimens of C. tranquebarica. which passed the late larval, prepupal, and pupal stages in a warm dry chamber, mean temperature 40° C. ; control, 4S7a'-e''W at 21 ° C. (Experiment 57) ; 458, the same as 457 but dry instead of moist; 4S&a'-b' control of the same (Experiment 58^: 461. the same moist warm treatment as described under 456 applied to C. purpurea .liiiibalis; 462, the same as 461 but dry in- stead of moist; for normal patterns see figure 512, plate XXVIII; a collected specimen from the same generation showing the e.xtreme type of cross band re- duction and forward curvature found either in the controls or the collections from the habitata ; 459, showing the pattern of a specimen of C. tranquebariia which was forced through its transformations in the fall by a temperature of Z7° C. beginning October I, so that there was no hibernation. This specimen was one emerged early in December. The others emerged in June but none of them showed any modification ; 46oa-&, the same treatment as 456 but dry instead of moist (Experiment 60) ; 463, showing the patterns of specimens of C. scutcl- laris lecontei Hald. subjected to conditions similar to those mentioned for figure 456; 463^ shows markings reduced below anything ever collected near Chicago or produced in the controls (Experiment 63) ; 464<;, b, c, showing the patterns of elytra of C. scutellaris lecontci subjected to mean temperature of 39° C. under very moist conditions (Experiment 64) ; 4651;. b, c, showing the elytral patterns of two specimens of C. scutellaris lecontci Hald, kept in an ice box during the pupal and prepupal stages; 10 to 12° C. from July 29 to September 3; 16 to 20° September 3 to October 16; 466 shows the elytron of a specimen kept at a mean temperature of 40° C, moist; 466a', b', c are the control of the same kept at 21° C. ; Ad'w' , x', y', z show elytra of specimens collected in the habitat from which the experimental material came, selected to show the range of variation ; 468a and w' . . a shows the middle band of a specimen of C. hirticollis showing the rounded angle, transverse portion perpendicular to the inner border of the elytron and the hooked portion at the eiid rounded — compare with the normal shown in 46&C1'. ILLINOIS lilOLOGIC.lL MONOGRAPHS VOLUME 3 4flf) EXPtKIMlMSb e rt^N^ ^' / ,/\«l ^^ ^0^^ <5 \^ uT-^ a'^^S d^^ c^^ d' ^56 CONTROL 5fi -, ''57 EXP58 458C(WTB0l 58 ""* ^9 ''^^ EXPERIMENT 60 '^'''tXPei ''^^EXpl^ CONTROLS EXPERIMtNTf,3 /;^-V7 c^^-Tl "^^ ^ 'lllb tXPtRIMENT h.S ,f?5 -5^ ^^.bJR^ ^^ ^^ y^ ^^ *"'' rnNTimi fiS _. '"'"cyoCfi 467 rnurimi ci ■ .^ ^~^ C0NTR0166 -> ''^'TXP66 ''6' CONTROL67- SI s^txpea SHELFORD COLORS OF TIGER liEETLES PLATE XX 112 ILLINOIS BIOLOGICAL MONOGRAPHS [506 PLATE XXI Explanation of Plate Figure 469. Showing the range of variation in the group of races included under the name tranguebarica Herbst. The classes of patterns are arranged in a series a, b, c, d, e, f, etc., from left to right and the percentage of individuals in each class for several localities is graphically represented. At the top is indi- cated the color of the elytra to which the patterns belong but these do not fall in the same classes as the patterns. The graphs arc numbered and the localities which they represent are numbered on figure 4690. The graphs are for the following localities, approximate, altitude, etc. Vegetation Altitude 2020 ft. 4S00 ft. 1060 ft. 2600 ft. 100+ ft. 2500 ft. 1500 ft. 200 ft. 1180 ft. 7536 ft. No. Locality 1. Las Vegas, Nev. 2. Provo, Utah o. San Bernardino, Cal. .(. Hagerman, Idaho 5. Galveston, Tex. (vicinity) 6. Dodge City, Kan. 7. Fayetteville, Ark. 8. Framingham, Mass. g. Aweme, Manitoba 10. Alamosa, Colorado The classes into which they are divided ar tlie curves are divided. No. Specimens and Climate 8 Desert IS " 10 Semi desert 5 " 130 Savanna 69 Steppe 4-^ Deciduous forest 1-49 " " 73 Steppe 7 somewhat arti icial and some of ILLISOIS BIOLOGICAL MONOGRAPHS VOLUME 3 O O o o o o S "=> SlIELFORD COLORS OF TIGER BEETLES PLATE XXI 114 ILLINOIS BIOLOGICAL MONOGRAPHS [508 PLATE XXII Explanation of Plate Figure 4691!. Showing the tlistrilnitioii of C. tranquebarica in N. America. The numbers refer to the graphs sliown in I'igure 469. The legend shows the elvtral color. ILUXOIS BIOLOGICAL MONOGRAPHS VOLUME 3 SHHLFORD COLORS OF TIGER BEETLES PLATE XXII ILLINOIS BIOLOGICAL MONOGRAPHS [510 PLATE XXIII Explanation of Plate Figure 470. Showing the range of variation in the group of races included under C. scuiellaris Say. General plan as in figure 469, plate XXI. Here the individuals are arranged into classes which are strictly geographic ; beginning in Massachusetts at the extreme left, they are arranged as encountered as one passes southward along the Atlantic coast and westward through the Gulf States. From Dallas, Texas, classes are arranged in order as one passes northward through western Oklahoma, Kansas, Nebraska, and South Dakota and tlien eastward through the Great Lakes. The classes on the extreme right (s and are from Aweme, Manitoba. No. Climate Locality Specimens Generation Collector Altitude or Vegetation I. Framingham, Mass 51 1902 1904 A. C. Frost 220 ft. Deciduous Forest 2. Providence, R. L 8S 1902 P.crt Nock soft. " 3- Aqueduct, N. Y. 98 1903 L. H. Joutel 50 ft. " " 4. Raleigh, N. C. 59 1904 C. S. Eriniley 320 ft. " ** s. Mobile, Ala. 20 191 1 V. E. Shelford 50 ft. It tt 6. Medora, Kan. 150 1904 " 1600 ft. Steppe 7- Topeka, Kan. 150 1904 " 900 ft. Savanna a. Elma, Iowa 30 1902 -1904 Rev. J. C. Warren 1000 ft. " 9- Starved Rock (Utica), 111. 40 1905 -1906 V. E. Shelford 470 ft. " 0, Miller, Ind. 200 1904 -1905 " 600 ft. Deciduous Forest t. Aweme, Man. ' N. Griddle 1 1 80 ft. Steppe n.f.rXOIS BIOLOGICAL MONOGRAPHS VOLUME 3 05 OO SllELFOKU COLORS OF TIGFR CEETLFS PMTE XXI 1 1 118 ILLIXOIS BIOLOGICAL MONOGRAPHS [512 PLATE XXIV Explanation ok Plate Figure 470a. Showing tlie distribution of the group of races included under C. sciitcUaris Say. The legend indicates the color of tlie elytron. The nunibers (italics) refer to the classes of color patterns indicated in figure 470, plate XXIII. The lines and numbers indicate mean annual rainfall in inches. The mean annual rainfall to the left or west of the line designated as 20 is less than 20 inthes, to the right or east more than 20 inches. To the east and south of the line desig- nated as 30 the mean annual rainfall is more than 30 inches. To the east and south of the line designated as 40 it is more than 40 inches. Note that the colors are fairly well correlated with rainfall. ILLIXOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 SHELFORD COLORS OF TIGER BEETLES PLATE XXIV 120 ILLINOIS BIOLOGICAL MONOGRAPHS [514 PLATE XXV Explanation of Plate Figure^ 471. Showing the range of variation in the group of races falling under C. purpurea Oliv. General plan of arrangement as in the preceding charts on other species (Pis. XXI and XXIII). In the case of this species the immacu- lated elytroned types which are very rare in occurrence are taken as a central type. Those to left are level ground inhabitants in which the reduction of pat- terns is characterized by a withdrawal of the middle band from the elytral mar- gin. Those to the right are the steep clay bank inhabitants, except possibly class "t" (C. decemnotata Say); classes a, b, c. C. ciinarrona Lee; d-h, C. purpurea Oliv., graminea Schpp., audobonii Lee, spreia Lee. Those to the right are splen- dida, Hentz, transversa Leng, denvcrensis Cas., limbalis Klg. The graphs are for the following localities with approximate altitudes, vegetation, etc. Climate Locality No. Specimens 1 Color or Race Altitude or Vegetation I. Fort Collins, Colo. 7 Green and black 5600 ft. Steppe 2. Framingham, Mass. 128 Winecolor, brown, some greenish 100 ft. Deciduous Forest 3- Puget Sound, Wash. 7 Green 10 ft. Conifer 4- Kimmich, Mo 29 trans'i'crsa 42s ft. Deciduous Forest S- Topeka, Kan. 100 splendida 900 ft. Savanna 6. Glencoe, 111. 54 limbalis 6coft. " 7- Aweme, Man. 10 lo.-^o ft. Steppe 8. Sedalia, Colo. Red Classes, p-s 5800 ft. " ILLINOIS BIOLOGICAL MOXOGRAPHS VOLUME 3 SHELFORD COLORS OF TIGER BEETLES PLATE XXV 122 ILLINOIS BIOLOGICAL MONOGRAPHS [516 PLATE XXVI Explanation of Plate FiGUKE 471a. Showing the distribution of the limbalis, denverensis, transversa, and 10 notata races of C. purpurea with numbers referring to the graphs in figure 471, plate XXV, and legend showing colors. ILLIXOIS BIOLOGICAL MOyOGRAPHS J-QLUME 3 SHELFORD COLORS OF TIGER BEETLES PLATE XXVl 124 ILLINOIS BIOLOGICAL MONOGRAPHS [518 PLATE XXVII Explanation of Plate Figure 472. Showing the distribution of the purpurea, graminea, audobonii, and cimarrona races of C. purpurea with numbers referring to the graphs on figure 471, plate XXV, and legend showing colors. ILLINOIS BIOLOGICAL MONOGRAPHS VOLUME 3 Q tfiye - BROWN pnrporea Oliv. aQuuTona Lee. ^BLACK Atidubomi Lee. □ RED X Cfl££,v (na=;cai Schpp. SHELFORD COLORS OF TIGER BEETLES PLATE XXVII 126 ILLINOIS BIOLOGICAL MONOGRAPHS [520 PLATE XXVIII Showing the parallelism of patterns of the north stem of W. Horn's phylogenj', and C. sexguttata Fabr. Compare the rows with one another. Explanation of Plate ■Figs. 473-474, tranquebarica Herbst subsp. plutonica Cas. (California) drawn from descriptions by Leng : 475-478, do. subsp. vibcx Horn (Las Vegas, Nev.) ; 479, do. (San Bernardino, Cal.) ; 480, greenish brown form of tranquebarica (Ha- german, Idaho) (roguensis Harris) ; 481, tenuicincta Sch. (Salt Lake) ; 482, tranquebarica (Framingham, Mass) ; 483, tenuicincta Schpp. (Saltair, Utah) ; 484, tranquebarica Herbst (Alamosa, Colo.) ; 485, do. (Las Vegas, Nev.) ; 486-490, scutellaris Say, varieties — see figure 468 and description ;, 491, echo Cas. (Great Salt Lake) ; 492, willistoni Lee. (Lake Como, Wyo.) ; 492, fulgida Say (Kansas) ; 494-496, latesignata Lee. (San Diego, Cal.); 497-501, pulchra Say (499-501 — .Al- pine, Texas, drawing supplied by Prof. H. F. Wickham, from specimens in his collection) ; 502, latesignata aber. tenuicincta Blaisdell (Saltair, Utah) ; 503-505, longilabris Say, varieties ; 504-505, (N. Mexico) ; 506-518, purpurea Oliv., varie- ties (see Fig. 470) ; 508-509, 516-518 (Sedalia, Colo.); 519, generosa subsp. man- toba Leng.; 522-523, sexguttata (Onaga, Kansas); 524, do. (Chicago); 525-526, do. (Woods Holl.) ; 527, sexguttata subsp. patruela Dej. (Lakehurst, N. J.) : 528, /.? guttata Dej. (Chicago); 529, ancosisconensis Harris; 530, repanda Dej. (Chi- cago); 531-532, generosa Dej.; 531, do. (Framingham, Mass.); 532, do. (Lake- hurst, N. J.) ; 533-534, venusta Lee. (Aweme, Man.) ; 535-536, limbata Say (Aweme, Man.) ; 536, purpurea, showing reduced and shortened marking. ILLIXOIS BIOLOGICAL MOXOGRAPLIS VOLUME 3 .ijl^ /r'^ SMELFORD COLORS OF TIGER BEETLES PLATE XXVIII 128 ILLIXOIS BIOLOGICAL MONOGRAPHS [522 PLATE XXIX Showing Development and General Modification of Colors in Experiments in C. scutellaris lecontei Hald. Figs. ; 538, 539' 540. 541 542. Figs. 543- 543 544 545 546 547 548. 549 550. 551. 55-. 553 554. 555. 556. 558. EXPLAN.\TI0N OF PlATE 542. Development of color in the ventral side. 4 hours after emergence. 10 hours after emergence. 11 hours after emergence. 15 hours after emergence — adult coloration. Adult coloration in a dark individual. 549. Color development and color changes in an individual of C lecontei. I hour after emergence. 1 1 hours after emergence. 13 hours after emergence; compare with 553. 15 hours after emergence. 3 to IS days after emergence ; drawn at end of third day. 42 days after emergence. 85 days after emergence. C. lecontei, color of modal class, Miller, Ind., April, 1906. C. lecontei. Miller, Ind., June, 1906. C. lecontei, color of modal class. Miller, Ind., .^pril, 1905. C. scutellaris rugifrons, typical specimen, Raleigh, N. C. C. scutellaris, typical specimen, Topeka, Kansas (not modal class). C. lecontei, larvae subjected to hot dry conditions during prepupal and pupal stages, note reduced markings and color — compare with normal ontogeny series above (Experiment 63) ; mean tempera- ture 37°; dry; compare with 554 and 553. C. lecontei, larvae forced by high temperature and brought through without hibernation (Experiment 591! ; mean temperature 37° C. ; moist. C. lecontei modified by cold conditions during the pupal stage; (E.xperi- ment 65) ; mean temperature. 12° C. ; moist. Note dull color and peculiarities of markings. Peculiar individual from Starved Rock (Utica), 111., showing the ten- dency for all the highly colored species to produce purple forms occasionally. This type occurs at Utica on the coarse sands. PLATE XXTX 542 ^ rt r 544 543 545 546 547 548 550 551 552 555 556 557 558 523] COLORS OF TIGER BEETLES— SHELFORD 129 PLATE XXX 130 ILLINOIS BIOLOGICAL MONOGRAPHS [524 Sliowing Color Development and General Modification in Experiments on Species Named. Explanation of Plate Figs. 559-562. Color development in Cicindela hirticollis. 559. Condition 4 hours after emergence. 560. Condition 15 hours after emergence. 561. Condition 21 hours after emergence. 562. Condition 21 days after emergence, full adult color. Figs. 563-565. Color development in C. purpurea. 563. Condition 20 hours after emergence. 564. Condition 4 days after emergence. 565. The same specimen as in figure 6, killed and dried on the fourth day after emergence. Figs. 565-570. Experimental modification of color and color pattern by condi- tions during the prepupal and pupa! stages. 566. Dwarfed specimen of C. hirticollis produced by forcing the larvae without hibernation in their last vifinter (Experiment 70) ; mean temperature, 37° C. ; moist. 567. Normal individual of C. tranquebarica, collected in the field. 568. Specimen with color modified by being kept in an ice box, during the pupal stage, like variety in eastern mountains (Experiment 6Sa) ; mean temperature, 12° C. ; moist. 569. Specimen with both pattern and color modified by hot dry conditions (Experiment 60); mean temperature, 37° C. ; dry. Like variety in the western states. 570. Specimen with both pattern and color modified by hot wet conditions, like variety in the moist southern states (Experiment 56) ; 37° C. ; moist. PLATE XXX 559 562 566 525] COLORS OF TIGER BEETLES—SHELFORD 131 PLATE XXXI 132 ILLINOIS BIOLOGICAL MONOGRAPHS [526 Showing Color Development and General Moditication in Experiments on C. purpurea subsp. limbalis. Explanation of Plate Figs. 571-574. Color development in C. purpurea subsp. limbalis. 571. Condition at emergence. Condition 12 hours after emergence. Condition 34 hours after emergence. Condition 15 days after emergence. Normal collected individual. Specimen killed and dried vthexi at stage shown in figure 573. Experimentally modified individual, in hot dry conditions during pre- pupal and pupal stages. Resembles specimens from eastern Kan- sas. Specimen with color modified by being kept in an ice box during the pupal stage (Experiment 656); mean temperature, 12° C. ; moist. Specimen modified by hot moist conditions during the prepupal and pupal stage (Experiment 61); mean temperature, 37° C. ; moist. 572 573 574- 575 576. 577. 579- f^ PLATE XXXI 57 572 573 574 575 576 I \J'^ 578 579 527] COLORS OF TIGER BEETLES— SHELFORD 133 PLATE XXXII 134 ILLIXOIS BIOLOGICAL MOXOCRAPHS [528 Explanation of Plate Figure 580. Showing the geographic distribution of types and patterns. The first series at the left are world-wide in distribution, being most generalized in Eurasia and North America. The second group of patterns belong to several groups of species but all are characterized by the presence of three spots at the base and along the elytral suture. They are most numerous in Africa and India. The next group shows the relatively rare type with the pattern oblique but in the opposite direction from the slope of the tip of the elytron. The last type is one showing peculiar joinings of markings characteristic of species found chiefly in Indo-Australian region. luj.voi': r.ioiociCAi. MoxncK.irir \-oi.VMr. 3 ^rzp SlIKLFORl) COLORS OF TIGER RRF.TI.F.S PLATF XXXII The I'opics of this vohiiiie \vci-i> distrihuted as follows: No. 1 Xoveiiiber 29, 1916 No. 3 May 5, 1917 No. 2 December 30. 1916 No. 4 June 30, 1917 Copies of tile title paji'e and table of eoiitents of Voliiiiie 111 iiuiy be .secured foi" binding from .Mr. II. H. ('unniii»liam. Administration l!uild- ing. University of Illinois, IVbana. 510, G m-6) I V. ^' Vol. Ill ILLINOIS BIOLOGICAL MONOGRAPHS August, 1916 No. I STUDIES ON THE FACTORS CONTROLLING THE RATE OF REGENERATION CHARLES ZELENY PRICE S1.2B Published by the University of Iixmois Under the Auspices of the Graduate School Urbana, Illinois UNIVERSITY OF ILLINOIS STUDIES PUBLISHED ILLINOIS BIOLOGICAL MONOGRAPHS The Illinois Biological Monographs is a serial published quarterly by the University of Illinois. It is the purpose of the publication to present in mono- graphic form research contributions of especial merit and of somewhat extended character. Each volume will include about 500 pages and 20 or more full page plates. The series may be obtained under a subscription price of thfee dollars yearly. Individual numbers will be sold separately at prices determined by the size and amount of illustration contained in a particular number. Vol, I Nos. I and 2. A revision of the Cestode family Proteocephalidae. With 16 plates. By G. R. La Rue. $2.00. No. 3. Studies on the Cestode family Anoplocephalidae. With 6 plates. By H. Douthitt, 80 cts. No. 4. Larval Trematodes from North American fresh-water snails. With 8 plates. By W. W. Cort. $1.20. Vol. II No. I. The classification of Lepidopterous larvae. With 10 plates. By. S. B. Fracker. $1.50. No. 2. On the Osteology of some of the Loricati. With 5 plates. By John E. Gutberlet. 60 cts. No. 3. Studies on Gregarincs, including twenty-one new species and a synopsis of the Eugregarine records from the Myriopoda, Coleoptera and Orthoptera of the world. With 1$ plates. By IMinnie E. Watson. $2.00. No. 4. The Genus Meliola in Porto Rico. With 5 plates. By Frank Lincoln Stevens. 75 cts. Vol. Ill No. I. Studies on the factors controlling llie rate of regeneration. By Charles Zcleny. $1.25. No. 2. T!ic head capsule and mouth-parts of Diptera. With 27 plates. By Alvah Peterson. (7); press.) UNIVERSITY OF ILLINOIS STUDIES IN LANGUAGE AND LITERATURE The Studies in Language and Literature are designed to include monographs in general linguistics and comparative literature; the classical languages and Sanskrit; the Romance languages; and English, the Scandinavian, and other Ger- manic languages. The title of the series will be so construed as to admit the publi- cation of such researches in the history of culture as may throw light upon the processes of language and the interpretation of literature. It is a serial published quarterly, for which the annual subscription price is three dollars. Vol. I Nos. I and 2. The phonology of the dialect of Aurland, Norway. By George T. Flom. $1.25.- Nos. 3 and 4. Studies in the Milton tradition. By Jolm Walter Good, $1.75. Vol II No. I. Thomas Warton : a biographical and critical study. By Clarissa Rina- ker. $1.00. No. 2. Illustrations of medieval romance on tiles from Chertsey Abbey. .B> Roger Sherman Loomis. 75 cts, " No, 3, Joseph Ritson : a critical biography. By Henry Alfred Eurd. Entered as second class matter July 27, 1915, at the rostolKcc at Url.ana, ] tj.e act of .August 24, 1912. UNIVERSITY OF ILLINOIS STUDIES IN THE SOCIAL SCIENCES The University of Illinois Studies in the Social Sciences are designed to afford a means of publishing monographs prepared by graduate students or members of the faculty in the departments of history, economics, political science, and sociology. Each volume will consist of about 600 printed pages annually. The subscription price is three dollars per year. Vol. r Nos. I and 2. Financial history of Ohio. By E. L..Bogart. $1.80. No. 3. Sources of municipal revenues in Illinois. By L. D. Upson. Out of print. No. 4. Friedrich Gentz : an opponent of the French revolution and Napoleon. By P. F. Reiff. Out of (>rinl. Vol. II No. I. Taxation of corporations in Illinois, other than railroads, since 1872. By J. R. Moore. 55 cts. Nos. 2 and 3. The West in the diplomatic negotiations of the American Revolution. By P. C. Phillips. $1.25. No. 4. The development of banking in Illinois, 1817-1863. By G. W. Dowrie. Out of print. Vol. Ill Nos. I and 2. The history of the general property tax in Illinois. By R. M. Haig. $1.25. No. 3. The Scandinavian element in the United States. By K. C. Babcock. $I.I5- No. 4. Church and state in Massachusetts, 1691-1740. By Susan M. Reed. $1.05. Aline Tn:;:itioii in tlic United States. V'\ T-, I-"-. Voutiy. (In prrss.) Vol. IV No. I. The Illinois Whigs before 1846. By C. M. Thompson. 95 cts. No. 2. The defeat of Varus and the German frontier policy of Augustus. By W. A. Oldfather and H. V. Canter. 75 cts. Nos. 3 and 4. The history of the Illinois Central railway to 1870. By H. G. Erownson. $1.25. Vol V \o. I. Inforcement of international law by municipal law in tlie United States. By Philip Quincy Wright. $1.25. Xn. 2. The life of Jesse W. Fell. By Frances M. Morehouse. 60 cts. Xo. 3. Land tenure in the United States with special rofcieiu-c to Illinois. By Charles L. Stewart. 75 cts. UNIVERSITY STUDIES Gfn'er.\l Sf.kies, Vol. I, II and III Partly out of print. A detailed list of these numbers will be sent on request. Vol. IV Xo. I The origin of the land grant of 1862 (the so-called Morrill act) and some account of its author, Jonathan B. Turner. By E. J. James. Cloth $1.35; paper, 75 cts. .Xo. 2. Fragment RA.sSC of Konongs Skuggsja, from an old Norwegian parch- ment codex of the Xlllth century, with heliotype copy and a study of the palaeography and the language. By G. T. Flom. $1.25. No. 3. Michel-Jean Sedaine: Le philosophe sans le sgavoir. Variorum critical edi- tion, based upon the second Paris edition of 1766 and its appendix, with an introduction treating of all known editions and translations. By T. E. Oliver. 5t.7.=I. PUBLICATIONS OF THE UNIVERSITY OF ILLINOIS Following is a partial list of the publications issued at the University : 1. The University of Illinois Studies in Language and Literature, Pub- lislieU quarterly. Three dollars a year. Address 158B Administration Building, University of Illinois. 2. The University of Illinois Studies in the Social Sciences. Monographs in history, economics, political science, and sociology. Published quarterly. Three dollars a year. Address 1588 Administration Building, University of Illinois. 3. The Illinois Biological Monographs. Published quarterly. Three dollars a year. Address isSB Administration Building, University of Illinois. 4. The University Studies. A series of monographs on miscellaneous sub- jects. Address Manager of University Studies, 1586 Administration Building, University of Illinois. 5. The Journal of English and Germanic Philology. Published quarterly. Three dollars a year. Address Manager of Journal of English and Germanic Philology, 158B Administration Building, University of Illinois. 6. The Bulletin of the Engineering Experiment Station. Reports of the research work in the Engineering Experiment Station. Address Director of Engi- neering Experiment Station, University of Illinois. 7. The Bulletin of the Agricultural Experiment Station. Address Di- rector of Agricultural Experiment Station, University of Illinois. 8. The Bulletin of the State Laboratory of Natural History. Address Director of State Laboratory of Natural History, University of Illinois. 9. The Bulletin of the State Geological Survey. Address Director of State Geological Survey, University of Illinois. ID. The Bulletin of the State Water Survey. Address Director of State Water Survey, University of Illinois. 11. The Report of the State Entomologist. Address State Entomologist, University of Illinois. 12. The Bulletin of the Illinois Association or Teachers of English. Address 301 University Hall, University of Illinois. 13. The Bulletin of the School of Education. Address 203 University Hall, University of Illinois. 14. The general series, containing the University catalog and circulars of special departments. Address The Registrar, University of Illinois. m