CROWN-GALL OF ALFALFA 
 
 (WITH PLATES VH-X) 
 
 O. T. WILSON 
 
 UNIVERSITY OF WISCONSIN 
 PH. D. THESIS \<\\ (n 
 
 Reprinted for private circulation from 
 The Botanical Gazette, Vol. LXX, No. i, July 1920 
 
( — 
 
 HBRARY OF CONGRESS 
 
 -:r:-^ivKn 
 
 APR 6 1922 
 
WfD CROWN-GALL OF ALFALFA 
 
 O. T. Wilson 
 
 (with plates vii-x) 
 
 Introduction 
 
 The crown-gall of alfalfa has been known in the United States 
 for only a few years, the first published record of its occurrence 
 in this country being in 1909. It was at that time observed in 
 California by Smith (32). Since then it has been reported from 
 Arizona (23), Oregon (26), and Utah (27). It is probable that 
 it also occurs in other western states. The disease has been known 
 somewhat longer in other countries. Patouillard and von 
 Lagerheim (28) in 1895 published the earliest record of its oc- 
 currence in connection with an outbreak in Ecuador. In 1902 
 Magnus (21) described the general features of the disease and 
 recorded its presence in Germany. It has also been reported 
 from England (29), and Italy, Sweden, and Switzerland (10). 
 
 The lack of detailed information makes it difficult to estimate 
 the economic importance of the disease. Old and powerful stands 
 were killed in the fields of Ecuador (19). Magnus considered 
 the disease serious at Colmar in Alsace. Salmon (29) advised 
 strict precautions against its spread in England. In the United 
 States McCallum (23) reported the disease as serious but not 
 widely distributed in Arizona at the time of his observations. 
 Smith noted it in only a few counties of California. According 
 to O'Gara (26), many plants two to seven years of age were de- 
 stroyed by the parasite in Oregon. In 191 5 resolutions were 
 adopted by the American Phytopathological Society (31) recogniz- 
 ing the serious possibihties of the disease and recommending its 
 investigation by the government. 
 
 It is the purpose of this study to contribute to our knowledge 
 of the life history of the causal organism. 
 
 Material and methods 
 Infected plants were secured from the vicinity of Medford, 
 Oregon. In the winter of 1913-1914 P. J. O'Gara sent specimens, 
 
 51] [Botanical Gazette, vol. 70 
 
52 BOTANICAL GAZETTE [july 
 
 and in the fall of the year 1914 material was received from 
 Dr. M. P. Henderson. Other material was secured from alfalfa 
 plants grown and infected in the greenhouse. 
 
 Standard histological and cytological methods were employed 
 for the examination of the material; reference to these will be 
 made in connection with various phases of the observations. 
 
 Observations 
 
 In 1902 Magnus (21) described the galls as branching tuber- 
 culate structures on the larger secondary roots of Medicago sativa. 
 Upon examining cross-sections of the galls, he found large brown 
 regions of irregular form, which proved to be cavities filled with 
 resting spores of the parasite. ''Thick- walled strongly en- 
 cysted mycelium" was found in many of the cavities, but he did 
 not find that the resting spores were attached to the hyphae. The 
 amount of mycelium present in the different cavities varied; it 
 was often entirely lacking. The hyphae were described as contin- 
 uous or branched, and the protoplasm of the host cells was often 
 completely displaced by these ''wandering hyphae." Magnus 
 thought that this mycehum might "awake to new life" after 
 the winter resting period. He described the resting spores as 
 spherical with one side flattened; he noted a colorless hyaUne 
 cell attached to the flattened side by means of a hyaline process. 
 Many pores were found in the centers of the flattened walls of 
 the spores. No other stages in the life of the parasite were men- 
 tioned by Magnus, and not all those described were figured. So 
 far as the writer has been able to learn, no subsequent work has 
 been published touching upon the life history of this organism, 
 which has been classified as Urophlyctis alfalfae (von Lagerheim) 
 Magnus. Figures of the galls have been published by various 
 writers (13, 22), and all agree with Magnus' original description. 
 
 The alfalfa plants which furnished the material for these ob- 
 servations were several years old. Numerous galls corresponding 
 to the structures described by Magnus and others as the crown- 
 gall of alfalfa were found upon the plants (fig. i). Free-hand 
 sections of such galls upon microscopic examination revealed 
 numerous brownish resting spores, like those figured by Magnus, 
 
i92o] WILSON—CROWN-GALL 53 
 
 occupying irregular cavities in the tissues of the gall (fig. 8). If 
 the spores are not scattered in sectioning they completely fill the 
 cavities. They average 40 /j, in diameter. When shaken from the 
 cavities they appear as a glistening brown powder. 
 
 With rare exceptions the spores are spherical in form, except 
 for the depression on one side (hg. 8). Great irregularities of 
 form may occur, especially near the borders of masses of spores, 
 where' the host tissues apparently interfere with the natural con- 
 formation of the spores. These irregular spores may conform to 
 the outlines of the host cell occupied (fig. 11). A group of slitlike 
 pores in the depressed surface is normally a prominent feature of 
 the spores (fi.g. 9). Bally (3) observed similar pores in the walls 
 of the spores of Urophlyctis Ruhsaameni. The walls of the 
 spores consist of two layers, the outer much heavier than 
 the inner (fig. 6). The outer layer is yellow-brown, glassy in 
 appearance, and brittle, as shown in sectioning. It is very re- 
 sistant to stains, this quality being characteristic of resting spores 
 of the Chytridiaceae. A positive reaction with phloroglucin indi- 
 cates that it is lignified, but it does not stain with safranin. The 
 inner layer of the wall is thin and hyaline in appearance ; it responds 
 to the zinc chloriodide test for cellulose. In this respect it is 
 like the wall of the sporange of Olpidium Viciae described by 
 KusANO (17). It lacks the rigidity and brittleness of the outer 
 layer. Ridges in the inner layer are of frequent occurrence (fig. 13) . 
 
 In the resting condition the protoplasm of the unstained spore 
 appears to be granular in nature and somewhat vacuolate (fig. 10). 
 The nuclei cannot be distinguished without staining. Flem- 
 ming's triple stain or Heidenhain's iron-alum-haematoxylin may 
 be employed to bring out the nuclei. Many small nuclei are 
 found to be scattered quite regularly through the cytoplasm of 
 the spores (fig. 15). It is difficult to detect the structure of the 
 nuclei at this stage; it becomes easier as the spore develops into 
 a sporange. There is considerable variation in the response of 
 the nuclei to stains. This is probably related to the difficulty in 
 obtaining very thin sections through the spores. Nuclei which 
 have been sectioned show that the chromatin is usually at the 
 periphery, often concentrated at one side. In some preparations 
 
54 BOTANICAL GAZETTE [julv 
 
 a nucleole appears clearly (fig. 24). The condition of the resting 
 spore at this stage corresponds to that found by Loewenthal (20) 
 in the sporange of Olpidium Dicksonii at the very beginning of 
 zoospore formation, when the presence of very numerous tiny 
 nuclei was noted. Before staining, the content of the resting 
 spore appears hyaline and refractive in section. Only the nuclei 
 stain deeply; about each nucleus is a clear region, the clear por- 
 tions being separated by lightly stained cytoplasm. 
 
 It is only in the resting stage that the heavy-walled bodies 
 should be called spores. From the nature of their further develop- 
 ment they are clearly potential sporanges. The resting spores of 
 the Chytridiaceae have frequently been called sporanges. 
 
 The development of the sporanges was first studied in Van 
 Tieghem cells, distilled water being the medium employed. The 
 spores undergo an immediate change when placed in water. The 
 granular appearance gives way to one in which the cell seems to 
 be filled with small globules of oil. This indicates the first changes 
 leading to the formation of zoospores. If the spores are crushed 
 at this stage, numerous globules of fat are freed which stain with 
 Sudan III or with osmic acid. As development continues, the 
 fat globules in the sporanges apparently become larger and less 
 numerous (fig. 14). This condition continues until the motile 
 zoospores are set free. 
 
 In water the wall of the sporange also undergoes an almost 
 immediate change; a swelling of the wall on that side takes place, 
 presumably because of the entrance of water through the pores 
 in the depressed surface. Busgen (5) observed a somewhat simi- 
 lar swelling of the wall in the sporanges of Cladochytrium Butomi. 
 The sporanges become almost or quite spherical in outline. The 
 pores become cracks, and after a time the outer layer may rupture 
 as the inner content becomes more turgid (fig. 18). Various 
 appearances are brought about by the protrusion of the inner layer 
 of the wall and its contents. Sometimes the outer layer is thrown 
 off like a cap, as Busgen observed in his study of Cladochytrium. 
 In the majority of cases, however, the outer layer remains intact. 
 The whole process indicates a softening or gelatinization of the 
 wall as compared with its previous brittle condition. 
 
19 20 WILSOX—CROWX-GALL 55 
 
 Just before the exit of the zoospores a motion of the sporange 
 contents is visible. The fatty globules are jostled about although 
 they do not flow together. In a spore which is not viable the fat 
 is frequently congregated into one or two large drops, but such 
 a spore was never seen to develop into a sporange. The motion 
 within the sporange is followed by the rupture or dissolution of 
 the inner wall layer, allowing the zoospores to escape through the 
 openings in the outer layer. The difficulty of observing their 
 direct exit is enhanced by the fact that the porous side of the spo- 
 range wall is almost always downward in the hanging drop. The 
 zoospores escape sometimes in groups, but usually singly. Often 
 they seem to have difficulty in locating the pores, and they may 
 swim about in the sporange for a considerable time or even dis- 
 integrate there (fig. 19). Atkinson (i) has observed an interesting 
 amoeboid movement of zoospores within the sporanges of Rhi- 
 zopkidium glohosum just previous to their escape. Although no 
 such amoeboid movements were seen at this stage in my material, 
 the alternation of resting and active periods is a comriion occur- 
 rence, just as was observed by Atkinson. 
 
 In stained sections the development of the sporanges may be 
 traced in a fairly complete series. Even in the very early stages 
 each nucleus seems to be related to a definite portion of the cyto- 
 plasm (fig. 26). The nuclei have never been observed to occur 
 in groups, but are rather uniformly distributed. As development 
 continues, the nuclei decrease in number, as shown by the number 
 present in a cross-section of a sporange. Apparently there is a 
 disintegration of many of the small nuclei, while those which are 
 to take part in the formation of zoospores increase in size. The 
 structure of the latter nuclei also becomes much more clearly dif- 
 ferentiated, and they stain with more uniformity (figs. 20, 27). 
 The chromatin is now aggregated in knots connected by slender 
 threads (fig. 16). In most of the preparations the connecting 
 threads are not easily seen; in such cases one or more knots are 
 apparent just within the nuclear membrane (figs. 20, 27); these 
 large nuclei are the centers of zoospore formation. Barrett (4) 
 found in Olpidio psis nuclei similar to those here described, which 
 served as centers of zoospore formation. The transition stages 
 
56 BOTANICAL GAZETTE [july 
 
 from the condition shown in fig. 15 to that in which the spores 
 are deHmited have not been clearly followed. Zoospore forma- 
 tion is evidently a very rapid process. With the possible excep- 
 tion of the cilia, the zoospores are quite fully developed before 
 their escape from the sporange. The vacuole and nucleus are 
 quite well defined (figs. 23, 25). Loewenthal (20) found a vacuole 
 and well defined nucleus in the zoospore of Olpidium Dicksonii 
 before its escape from the sporange. Barrett observed the vacuole 
 in the zoospore of Olpidiopsis, but apparently did not succeed in 
 staining the nucleus. A cleavage of the sporange cytoplasm pre- 
 vious to zoospore formation is apparent in well stained sections. 
 This cleavage begins at the margin of the sporange, and works 
 toward the central region (fig. 22). The process apparently cor- 
 responds to that described by Harper (16) for Woroninella. 
 
 An interesting variation from this development is of quite 
 common occurrence, but has been observed only in fixed and 
 stained material, probably because it is masked in fresh material 
 by peripheral fat globules in the sporange. Soon after the nuclei 
 begin to enlarge in the stages preceding zoospore formation, one 
 of them (sometimes several) undergoes an especially rapid develop- 
 ment (figs. 17, 21), becoming separated from the other nuclei by 
 a surrounding portion of denser cytoplasm. This is often quite 
 clearly shown, even in unstained sections. A spore may be clearly 
 delimited about this nucleus, while the other spores are yet in an 
 early stage of development. This spore is much larger than the 
 others formed in the same sporange. All the zoospores are alike 
 in structure, the only marked variation being in size. The forma- 
 tion of several of the large spores in the same sporange is excep- 
 tional (fig. 29). 
 
 When freed in water, the zoospores exhibit a great variety of 
 movement. Often the long cilium, which is quite clearly visible, 
 seems to impede the movement of the zoospore, which exhibits 
 a violent jerking motion. Irregular gyrations are common. Fre- 
 quently a zoospore moves back to the sporange from which it 
 escaped and seems to seek an entrance. At times the zoospores 
 move very rapidly from the field of observation in a direct line. . 
 The movements become less vigorous after a time; the ellipsoid 
 
i92o] WILSON— CROWX-GALL 57 
 
 or ovoid changes to a spherical form, and the long cilium becomes 
 more plainly visible than before, dragging behind passively. This 
 last described feature has been reported for the motile spores of 
 various Chytridiaceae. Periods of activity alternate with periods 
 of passivity, during which the vacuolated condition is very evident. 
 The fat drop is visible only during the active periods. Barrett 
 (4) and Butler (6) have observed pulsations of the vacuoles in 
 the spores of Olpidiopsis at this stage. 
 
 The fat drop is the most prominent internal feature of the 
 active zoospore. It may be seen to shift rapidly when the zoo- 
 spore is in motion. After a period varying from a few minutes 
 to several hours, terminated by sluggish amoeboid movements, 
 the zoospore comes to a final rest and soon disintegrates. It 
 seems likely that the refractive body so commonly observed in 
 the zoospores of the Chytridiaceae is a drop of fat occupying a 
 vacuole. Atkinson (2) observed the presence of a prominent fat 
 drop in the zoospore of Rhizophidiiim brevipes. 
 
 Amoeboid movements of zoospores have frequently been men- 
 tioned by investigators of the Chytridiaceae. Schenk (30) ob- 
 served this phenomenon as early as 1858. Dangeard (8) noted 
 that the zoospores of Chytridium xylophilum creep like amoebae. 
 BtJSGEN (5) observed similar movements in his study of Clado- 
 chytrium Butomi. 
 
 I did not directly observe the exit of the large zoospores from 
 the sporanges, but they were seen in considerable numbers moving 
 slowl}^ about in the water, remaining close to the sporanges from 
 which they had emerged. The content of these large zoospores 
 appears to be more granular than that of the smaller ones, which 
 latter appear hyaline in water. Very soon after their exit the 
 large zoospores are surrounded by the smaller ones. One or several 
 of the latter move swiftly toward a large zoospore and become 
 attached to it. As many as five small zoospores have been seen 
 adhering to one large spore, but in all cases only one remains at- 
 tached. There seems to be no uniformity as regards the point of 
 attachment. The small zoospore which remains attached loses 
 its cilium just at the time of contact. Kusano (17) reported a 
 resorption of the cilium at this stage in Olpidinm. After a small 
 
58 BOTAXICAL GAZETTE [july 
 
 zoospore has become permanently attached to a large one, the 
 latter continues to move about for a time before coming to rest. 
 An amoeboid form is finally assumed, and disintegration soon 
 follows. 
 
 Fusion of zoospores has been reported but rarely in the Chy- 
 tridiaceae. Probably this is because of the very limited obser- 
 vations upon the motile stages. Fisch (14) observed and figured 
 a fusion of zoospores for Reesia amoeboides. Dangeard (9) noted 
 an apparent fusion in Sphaerita endogena. Kusano observed a 
 clear case of fusion in Olpidium, the fusing cells being similar in 
 size. The observations of Sorokin (33) upon Tetrachytrium, of 
 Atkinson (2) upon Lagenidium, and of von Lagerheim (18) upon 
 Rhodochytrium should also be mentioned in this connection. 
 
 The motile spores were fixed and stained upon the cover slip 
 by the use of osmic acid and gentian violet. Each spore was found 
 to have a short cilium as well as the long one visible in the water 
 (fig. 28). The motion of the zoospore, after the long ciHum be- 
 comes passive, is probably due to the activity of the shorter cilium. 
 It seems possible that the uniciliate condition is not so common 
 among the Chytridiaceae as has been thought. Care in staining 
 and observation is necessary for a successful determination of the 
 number of ciha. Cornu (7) found only one cilium borne by the 
 motile spore of Olpidiopsis. Fischer (15) later found two, which 
 observation was confirmed by Barrett (4) . Cilia of unequal length 
 on the same zoospore have been reported and illustrated for various 
 members of the family, as for example Sphaerita endogena (9) . 
 
 The cilia of the motile cells of the organism under consideration 
 are attached at the same end of the zoospore to what seems to 
 be a plateUke thickening of the membrane. The manner of attach- 
 ment is much more clearly seen in the larger zoospores (figs. 31,32). 
 The oil vacuole is near the place of attachment of the cilia (fig. 40). 
 The nucleus lies back of the vacuole, imbedded in the cyto- 
 plasm. From the ciha to the nucleus there is a connecting, cone- 
 shaped, apparently fibrous structure, which extends through the 
 vacuole (figs. 31, 32}. Fig. 25 shows a general similarity to the 
 zoospore structure described by Loewenthal (20) for Olpidium, 
 and figured by Nowakowski (25) for Polyphagus Euglenae. 
 
i92o] WILSON— CROnW-GALL 59 
 
 Fusion stages may be quite clearly followed in material fixed 
 and stained on the slide. Fig. 28 shows two zoospores, a large 
 one and a small one, apparently in the same condition as when 
 freed from the sporange (see fig. 30). The nuclei of stained zoo- 
 spores often appear to be within a vacuole because of their position 
 when fixed to the slide. The larger zoospore, either before or at 
 the approach of the smaller one, may put out one or more pro- 
 jections in the direction of the latter (fig. 33). Fusion may take 
 place at the apex of such a projection (figs. 34, 36). The two 
 nuclei may be seen within the larger body following the fusion 
 (fig. 35). The fusion of the nuclei has not been observed. Figs. 
 37 and 38 possibly indicate stages following a fusion succeeded by 
 a division of the fusion nucleus. Many small nuclei appear in 
 the body of the organism at this stage (fig. 37), and there is some 
 evidence of a cell multiplication by budding (figs. 38, 44, 51). It 
 is entirely possible, however, that a multiplication of the nuclei 
 in either the large or the small motile bodies may occur without 
 any nuclear fusion. The observations at this point do not justify 
 very definite conclusions, since the staining of the zoospores in 
 toto makes the following of the nuclear phenomena decidedly diffi- 
 cult. There seem, however, to be some grounds for maintaining 
 that the phenomena just described constitute a true case of heter- 
 ogamy. One cell is characterized by a large body, slow movements, 
 and a nucleus of average proportional size. On the other hand, 
 the smaller cells are swift of movement and have nuclei very large 
 in proportion to their size. The oil drop probably serves as reserve 
 food for the temporary nutrition of either gamete in case fusion 
 does not occur, or of the zygote in case it does occur. 
 
 In order to secure the development of the zoospores upon the 
 host, young alfalfa seedhngs were carefully washed and introduced 
 into small vessels containing sterile distilled water, so that only 
 the roots and the low^er parts of the stem were immersed. A gall 
 from an infected plant was carefully washed, crushed, and intro- 
 duced into the water. After one day the seedhngs were removed 
 from the water, crushed under cover shps on slides, and examined 
 for evidence of the presence of the parasite. On and in the tissues 
 of the host which had been at the water level were numerous 
 
6o BOTANICAL GAZETTE [july 
 
 amoeboid bodies, as well as zoospores that still retained their 
 characteristic form. The ciha, however, were no longer to be seen. 
 The amoeboid bodies were watched for some hours and were seen 
 to become clustered and to move in masses; they could also be 
 seen to bud at times, much as in the case of the ciliated cell shown 
 in iig. 38. Budding of the amoebae in the infection stage has 
 been reported to occur in Plasmodiophora (11). Seedlings left 
 under these conditions developed small galls at the bases of the 
 secondary roots. 
 
 To secure stained preparations of the plasmodium in its early 
 stages, young alfalfa seedlings were grown in a pot in which a 
 badly infected mature plant was also growing. At various times 
 seedlings were removed and examined. Small galls were soon 
 found at the bases of the secondary roots of some of the seedlings. 
 As soon as the plants began to produce crown buds, galls appeared 
 upon the crown (fig. 3). These young galls were fixed, imbedded, 
 sectioned, and examined after various staining processes. The 
 Plasmodium of the parasite was found to be widely extended in 
 the galls. An amoeboid or plasmodial vegetative condition of the 
 parasite within the host has often been noted in members of 
 the Chytridiaceae. Fischer (15) and Barrett (4) have observed 
 it in species of Olpidiopsis. Cornu (7) observed it, as well as 
 the amoeboid movement of the zoospores, in various members of 
 the family. He also noted a suggestion of cleavage in what he 
 called the plasmodium of Rozella, and suggested its formation by 
 the union of many amoeboid zoospores. Fischer went so far as 
 to classify several genera upon the basis of differences in the Plas- 
 modia. He observed that the protoplasm of the parasite mingled 
 intimately with that of the host and gradually displaced it. Fisch's 
 (14) description of Reesia amoeboides is a striking suggestion of 
 the near relationship of the group to the Myxomycetes. A second- 
 ary infection of cells occupied by the plasmodium is of common 
 occurrence. As a result, the cells of the galls are often occupied, 
 not only by the plasmodium. but also by the earher stages of 
 the parasite (figs. 39, 41). 
 
 The nuclei in the plasmodium are very numerous, corresponding 
 quite closely in size and appearance to those in the resting spores 
 
1920] WILSON CROWN -GALL 6 1 
 
 (ligs. 42, 43. 45). The central c'ear portion surrounded by periph- 
 eral chromatin is characteristii^ of their appearance, although in 
 many cases the whole nucleus lakes up the chromatin stains. The 
 cytoplasm of the plasmodium is slightly granular and stains very 
 lightly. It has not yet been possible to keep the plants infected 
 in the greenhouse long enough to secure the development of the 
 resting spores within the galls. 
 
 In order to examine the plasmodia in older galls, sections were 
 stained with the triple stain and with haematoxylin. Only com- 
 paratively young galls were sectioned, as considerable foreign 
 mycelium was found to be present in the older galls. Infection 
 by other fungi, and the consequent presence of a foreign mycelium 
 which has no connection with the organism causing the galls, is 
 to be expected because of the contact of the infected parts with 
 the soil. Sections of smooth intact galls showed no such signs 
 of foreign contamination. The same type of plasmodium, however, 
 that was found on the exterior of seedlings exposed to infection, 
 and within the tiny galls formed upon seedlings grown in infected 
 soil, was also found throughout the cells of the galls containing 
 resting spores. In marginal cells of some galls the amoeboid cells 
 were found in numbers (fig 47), preceding the formation of a plas- 
 modium such as is shown in fig. 5. 
 
 In some cases the plasmodium forms an irregular streaming 
 mass, forcing its way through and between the cells of the gall 
 tissue (fig. 45). In other cases it ramifies through the cells as a 
 network of naked protoplasm (fig. 42). MultipHcation of the 
 Plasmodia may occur in these stages by budding and fragmenta- 
 tion (fig. 51). The walls of the host cells are often dissolved or 
 gelatinized in advance of the main body of a plasmodial mass, 
 presumably by enzymatic action (fig. 4). 
 
 The resting spores are formed in cavities or pockets occupied 
 by the parasite. Remains of host walls are mingled with the plas- 
 modial masses of the parasite, and a clear staining of the material 
 at this stage is almost unattainable. A yellow coloration pervades 
 the unstained content of the pockets previous to the formation 
 of the spores. At the time of spore formation the yellow colora- 
 tion is limited to the outer walls of the spores, their content being 
 
62 BOTAXICAL GAZETTE [july 
 
 clear and refractive. Just how the formation of the spores takes 
 place has not been determined. 
 
 Around the margins of the larger cavities containing spores the 
 naked protoplasm of the plasmodium may be found in a more or 
 less shrunken condition. The similarity of this protoplasm to 
 that within the spores is convincing evidence of the origin of the 
 spores from the plasmodium, the most noticeable difference being 
 in the more regular arrangement of the nuclei within the spores. 
 Isolated host cells or groups of cells may be occupied individually 
 by separate plasmodia; it would seem that these small plasmodia 
 may develop walls about themselves and become resting spores. 
 The subject of spore formation is in need of very careful and pro- 
 longed investigation. 
 
 Primary infection, the actual entrance of the parasite into an 
 uninfected host, has not yet been observed. The fact that the 
 parts invaded are the adventitious buds and the secondary roots 
 in the very earliest stages of development minimizes the chances 
 of such an observation. There is little doubt, however, that 
 either the zoospores before fusion, the zygotes, or the plasmodia 
 formed on the surface of the host may penetrate the embryonic 
 host tissues. In young bud galls cases have been found in which 
 practically every cell of the growing tip was occupied by the para- 
 site (fig. 7). From this figure it will be seen that the infecting 
 cells of the parasite vary greatly in form and size after penetration. 
 Often they are found in contact with the host nucleus, the latter 
 being still intact. Sometimes a host cell contains several of the 
 invading cells. Sometimes a fusion of a number of these invading 
 cells, preparatory to the formation of a larger plasmodium and the 
 breaking down of the separating host cell walls, may be noted. All 
 these features are illustrated in fig. 7. 
 
 The discussion of the studies upon the life history of the para- 
 sitic organism may be concluded by a brief reference to the subject 
 of nuclear divisions in the different stages. Had division figures 
 not been found, identification of the nuclei in the plasmodium 
 would have been questionable, since these nuclei are very small 
 and details of their structure are not easily distinguished. Divi- 
 sion figures, however, are very common in some of the prepara- 
 
i92o] WILSON— CROW X -GALL 63 
 
 tions, and are easily recognized as such upon careful examination. 
 The process of division is clearly mitotic (figs. 46, 48-50, 53-57). 
 Dark bodies suggesting centrosomes are commonly visible in the 
 metaphases at the poles of the spindles (figs. 46, 50). The early 
 divisions of the nuclei in the amoeboid infection stages have been 
 followed with only partial success. The figures correspond to 
 those seen in the later divisions, but average considerably larger 
 (fig. 44). The centrosomes, if the dark bodies at the poles of the 
 spindles are to be called such, are a constant feature in good prepa- 
 rations. Clearly recognizable division figures within the spo- 
 range have been found less frequently. There is some evidence 
 that a division occurs while the nuclei are yet very small (fig. 15), 
 and much better evidence of a division just preceding the forma- 
 tion of zoospores (figs. 17, 48). Whether these are in the nature 
 of reduction divisions must be left for future determination. Care- 
 ful study has failed to lead to a definite conclusion as to the chro- 
 mosome number. In the divisions immediately preceding the 
 formation of the zoospores the number of chromatic bodies seen 
 is apparently four (fig. 48). In all the division figures observed, 
 at whatever stage, the spindle seems to be intranuclear, corre- 
 sponding to the findings of other investigators of members of the 
 Chytridiaceg-e. 
 
 Classification of parasite 
 
 VonLagerheim (19, 28), who first noted the occurrence of the 
 crown-gall of alfalfa, classed it in the genus Cladochytrium with 
 the specific name alfalfae. Magnus (21), in his article of 1902, 
 gave strong reasons for removing the organism from that genus 
 and referred it to the genus Urophlyctis, retaining the specific name 
 given by von Lagerheim. The terminology Urophlyctis alfalfae 
 (von Lagerheim) Magnus has been generally accepted in later 
 works. Although Magnus was right in removing the parasite 
 from the genus Cladochytrium, it is doubtful whether he" was justi- 
 fied in placing it in the genus Urophlyctis, on the basis of his limited 
 observations. The description of the genus Urophlyctis as given 
 in Saccardo's Sylloge Fungorum (7: p. 303) is as follows: 
 
 Urophlyctis Schroet. Krypt. Fl. Schles. Pilze. p. 197. (Etym. oura Cauda et 
 phlyctish\x]l?i..) Zoosporangia sessilia in plantis vivis, fasciculis tilamentorum 
 
64 BOTAXICAL GAZETTE [july 
 
 immersa. Sporangia perdurantia intra cellulas plantarum viventium e 
 mycelio filiformi perforatas formata, copulatione cellularum similium orta. 
 Cellulae alterius protoplasma in alteram effunditur, haec vero crescit et mem- 
 brana crassa cincta in sporangium perdurans mutatur. 
 
 The studies of the alfalfa crown-gall organism by Magnus did 
 not establish the characters cited by Saccardo. The descriptions 
 of the genus as given by Engler and Prantl (12) and by Migula 
 (24) also include characters that Magnus apparently did not verify. 
 
 The writer has failed to observe, in the organism studied, 
 the characters ascribed to the genus Urophlyctis. In view of the 
 incompleteness of the results, no change of classification is sug- 
 gested at the present time. It seems highly desirable, however, 
 that careful investigations be made of the various organisms referred 
 to this genus. Possibly it may prove necessary to discard some 
 of the previously used diagnostic characters and to redescribe 
 the genus on the basis of fuller observations. The relationships of 
 the Chytridiaceae are in an unsettled state. There have been 
 many suggestions in the literature that would lead one to ques- 
 tion whether or not this family finds its proper place among the 
 Phycomycetes. The studies upon which classification has been 
 based in many cases have been very superficial, and few eft'orts 
 have been made to follow out complete life histories. Cases in 
 which an amoeboid or plasmodial stage has been noted, with the 
 absence of anything resembling mycelium excepting naked threads 
 of protoplasm, furnish reason to suspect that the family is more 
 closely related to the Myxomycetes than to the Phycomycetes. 
 
 Summary 
 
 1. The resting spores, placed in water cultures, develop into 
 sporanges. 
 
 2. Within these sporanges are formed motile zoospores of two 
 sizes; frequently one large zoospore and many small ones are formed 
 in the same sporange. 
 
 3. One or several small zoospores may become attached to one 
 large zoospore. Only one remains permanently attached. There 
 is some evidence that this attachment is related to a sexual fusion. 
 
 4. The movement of the large zoospore continues after the 
 attachment of the small one. 
 
1920] WILSON— CROWN-GALL 6$ 
 
 5. The small zoospores, the large zoospores, and the united 
 zoospores (zygotes ?) become amoeboid after a period of motility. 
 
 6. In the amoeboid state, singl}" or in groups, these bodies may 
 be observed upon the surface of the host. 
 
 7. In infected soil young alfalfa seedlings develop galls in which 
 Plasmodia are found. 
 
 8. In older galls similar plasmodia are found, which ramify 
 through the tissues of the gall. 
 
 9. The resting spores are formed in cavities within the tissues 
 of the galls. 
 
 10. The cytoplasmic and nuclear contents of the resting spores 
 in the dormant condition correspond to those of the plasmodium 
 in the stage immediately preceding the formation of resting spores. 
 
 University of Cincinnati 
 Cincinnati, Ohio 
 
 LITERATURE CITED 
 
 1. Atkinson, G. F., Intelligence manifested by swarm-spores of Rluzophi- 
 dium globosum. Box. Gaz. 19:503-504. 1894. 
 
 2. — , Some fungous parasites of algae. Bot. Gaz. 48:321-338. 1909. 
 
 3. Bally, W., Cytologische Studien an Chytridineen. Jahrb. Wiss. Bot. 
 50:95-153. 1911. 
 
 4. Barrett, J. T., Development and sexuality of some species of 01 pidi op- 
 sis. Ann. Botany 26:209-238. 1912. 
 
 5. BtJSGEN, M., Beitrag zur Kenntnis der Cladochytrieen. Cohn's Bei- 
 trage zur Biologic der Pflanzen 4:269-283. 18S7. 
 
 6. Butler, E. J., An account of the genus Pythium and some Chytridiaceae. 
 Memoirs Dept. Agric, India, Botanical Series 1:1-160. 1907. 
 
 7. CoRNU, M., Chytridinees parasites des Saprolegniees. Ann. Sci. Nat. 
 Bot. V 15:112-198. 1872. 
 
 8. Dangeard, p. a., Recherches sur les organismes inferieurs. Ann. Sci. 
 Nat. Bot. VII 4:241-341. 1886. 
 
 9. , Memoire sur les parasites du noyau et du protoplasm. Le Bo- 
 
 taniste 4:199-231. 1894-1895. 
 
 10. Delacroix, G., and Maublanc, A., Maladies des plantes cultivees. 
 
 11. Maladies parasitaires, p. 78. Paris. 1909. 
 
 11. DuGGAR, B. M., Fungous diseases of plants, p. 140. 1909. 
 
 12. Engler, a., and Prantl, K., Die naturlichen Pflanzenfamilien, i': 
 Leipzig. 1897. 
 
 13. Eriksson, J., Fungoid diseases of agricultural plants. JMolander's trans- 
 lation, p. 30. 191 2. 
 
 \ 
 
66 BOTANICAL GAZETTE [july 
 
 14. FiscH, K., Beitrage zur Kenntnis der Chytridiaceen. Berlin. 1884. 
 
 15. Fischer, A., Untersuchungen iiber die Parasiten der Saprolegnieen. 
 Jahrb. Wiss. Bot. 13:286-371. 1882. 
 
 16. Harper, R. A., Cell division in sporangia and asci. Ann. Botany 
 13:467-525. 1899. 
 
 17. KuSANO, S., On the life-history and cytology of a new Olpidium. Jour. 
 Coll. Agric. Imperial Univ. Tokyo 4:141-199. 1912. 
 
 18. VON Lagerheim, G., Rhodochytrium nov. gen., eine Uebergangsform von 
 den Protococcaceen zu den Chytridiaceen. Bot. Zeit. 51:43-51. 1893. 
 
 19. , ]Mykologische Studien. I. Beitrage zur Kenntnis der parasitischen 
 
 Pilze, I. tJber eine neue Krankheit der Luzerne {Medicago sativa L.). 
 Bihang K. Svenska Vet. Akad. Hand. 24: no. 4. 1898. 
 
 20. LoEWENTHAL, W., Wcitcre Untersuchungen an Chytridiaceen. Archiv 
 fiir Protistenkunde 5:221-239. 1905. 
 
 21. Magnus, P., tJber die in den knolligen Wurzelauswuchsen der Luzerne 
 lebende Urophlyctis. Ber. Deutsch. Bot. Gesells. 20:291-296. 1902. 
 
 22. jNIassee, G., Diseases of cultivated plants and trees. London. 1910. 
 
 23. IMcCallum, W. B., Report of the botanist. Ariz. Agric. Exp. Sta. Rept. 
 20:583. 1909. 
 
 24. MiGULA, W., Kryptogamen-Flora von Deutschland, Deutsch-Oesterreich 
 and der Schweiz, 3': Gera. 1910. 
 
 25. NowAKOWSKi, L., Polyphagus Euglenae. Cohn's Beitrage zur Biologie 
 der Pfianzen 2:201-219. 1877. 
 
 26. O'Gara, p. J., Urophlyctis alfalfae, a fungous disease of alfalfa occurring 
 in Oregon. Science N.S. 36:487-488. 1912. 
 
 27. , Existence of crown-gall of alfalfa caused by Urophlyctis alfalfae 
 
 in the Salt Lake Valley, Utah. Science N.S. 40:27. 1914. 
 
 28. Patouillard, N., and Lagerheim, G., Champignons de I'Equateur. 
 Bull. Herb. Boissier 3:53-74. 1895. 
 
 29. Salmon, E., Urophlyctis alfalfae, a fungous disease of lucerne in England. 
 Gardener's Chronicle 39:122-123. 1906. 
 
 30. SCHENK, A., liber das Vorkommen contractiler Zellen im Pflanzenreiche. 
 Wiirzburg. 1858. 
 
 31. Shear, C. L., Report of the secretary-treasurer, sixth annual meeting 
 of the American Phytopathological Society. Phytopathology 5:128-132. 
 
 1915- 
 
 32. Smith, E. H., A note on Urophlyctis alfalfae (v. Lagerh.) P. Magnus in 
 California. Science N.S. 30:211-212. 1909. 
 
 33. SoROKiN, N., Einige neue Wasserpilze. Bot. Zeit. 32:305-315. 1874. 
 
 EXPLANATION OF PLATES VII-X 
 With the exception of plate VII, the figures were drawn with the aid of 
 an Abbe camera lucida at table level. Leitz oculars and objectives were 
 used, giving the magnifications indicated. 
 
I920J WILSOX— CROWN -GALL 67 
 
 PLATE VII 
 
 Fig. I. — Alfalfa plant badly infected with crown-gall; X|. 
 
 Fig. 2. — Single galls; Xi. 
 
 Fig. 3. — Seedling infected in greenhouse, galls beginning to develop at 
 crown; Xi. 
 
 Fig. 4. — Section showing action of parasite upon walls of host cells in 
 advanced stage ; parenchyma tissue ; X300. 
 
 Fig. 5. — Plasmodium on border of gall ; X300. 
 
 Fig. 6. — Sections of resting spores showing outer and inner layers of 
 wall; X130. 
 
 Fig. 7. — Section of young gall on seedling showing cells of parasite in 
 host cells; some of former fused into tiny plasmodia, others in contact with 
 host nuclei; X300. 
 
 Fig. 8. — Section of gall showing cavities and resting spores of parasite. 
 
 PLATE VIII 
 
 Fig. 9. — Unstained resting spore, view of hollowed surface; X750. 
 
 Fig. 10. — Unstained resting spore, side view; X750. 
 
 Fig. II. — Resting spore conforming to host cell in which it has developed; 
 X750. 
 
 Fig. 12. — Unstained resting spore showing marginal vacuoles; X750. 
 
 Fig. 13. — ^Irregular relation of w^alls of resting spore ; X1800. 
 
 Fig. 14. — Unstained sporange when first put into hanging-drop culture; 
 X750. 
 
 Fig. 15. — Section of resting spore; may be stage just following resting 
 period; X1800. 
 
 Fig. 16. — Group of nuclei in sporange; chromatin in marginal aggrega- 
 tions; X1800. 
 
 Fig. 17. — Division figures in sporange, apparently just preceding zoo- 
 spore formation; large nucleus destined to be nucleus of large zoospore; X 1800. 
 
 Fig. 18. — Spore formation about completed, unstained; note rupture of 
 outer w-all; X750. 
 
 Fig. 19. — Unstained sporange containing a few zoospores; most of those 
 formed already escaped through prominent opening; X7S0. 
 
 Fig. 20. — Section of sporange just preceding formation of zoospores; 
 X1800. 
 
 Fig. 21. — Section of sporange showing large nucleus which will be included 
 in large zoospore ; X1800. 
 
 Fig. 22. — Marginal portion of sporange showing beginning of cleavage 
 into zoospores; X1800. 
 
 Fig. 23. — ^Zoospores almost fully formed in sporange; X 1800. 
 
 Fig. 24. — Nuclei of sporange showing nucleoles ; X1800. 
 
 Fig. 25. — Zoospore just before exit from sporange; X1800. 
 
 Fig. 26. — Nuclei of resting condition of spore. 
 
68 • BOTANICAL GAZETTE [july 
 
 Fig. 27. — Section of sporange dividing into zoospores; note characteris- 
 tic arrangement of chromatin in knot at one end or side of nucleus; X1800. 
 
 PLATE IX 
 
 Fig. 28. — -Free zoospores of two sizes; X1800. 
 
 Fig. 29. — Group of large zoospores formed in same sporange; X1800. 
 
 Fig. 30. — Large and small zoospore within same sporange; X1800. 
 
 Figs. 31,32. — Two large zoospores; X1800. 
 
 Fig. 33. — ^Large and small zoospore showing projection of former toward 
 latter; position may be accidental; X1800. 
 
 Figs. 34, 36. — x\ttachment of small zoospores to large; X1800. 
 
 Fig. 35. — Binucleate zygospore following fusion (?); X1800. 
 
 Fig. 37. — Amoeboidal stage following fusion, or perhaps developing with- 
 out fusion; nuclei have multiplied; X1800. 
 
 Fig. 38. — Apparent budding in amoeboid stage; X1800. 
 
 Figs. 39, 41. — Plasmodium and young infecting amoebulae in same host 
 cells; X1800. 
 
 Fig. 40. — Free zoospore, small size; X1800. 
 
 PLATE X 
 
 Fig. 42. — Plasmodium spreading through tissue of host; host nucleus 
 visible; X1800. 
 
 Fig. 43. — Plasmodium breaking through wall of host cell; X1800. 
 
 Fig. 44. — Amoebulae within tissues of host; note nuclei in division; 
 X1800. 
 
 Fig. 45. — Note as for fig. 42 ; no host nucleus visible. 
 
 Fig. 46. — Nuclei of Plasmodium in division; X1800. 
 
 Fig. 47. — ^Amoebulae massed in marginal cells of gall; X1800. 
 
 Fig. 48. — Division figures in sporange ; X1800. 
 
 Fig. 49. — Prochromosomes ( ?) in nuclei of sporange; X1800. 
 
 Figs. 50,53. — Anaphases in Plasmodium; X1800. 
 
 Fig. 51. — Budding or fragmentation of Plasmodium; X1800. 
 
 Fig. 52. — Nuclei of sporange just preceding formation of spores; X 1800. 
 
 Figs. 54-57. — Division figures in Plasmodium within host; X1800. 
 
BOTANICAL GAZETTE, LXX 
 
 PLATE VII 
 
 7 
 
 -» 
 
 ,' 
 
 
 
 p=^-' 
 
 * 
 
 H 
 •' 
 
 
 M . : • ^|5»4£ k 
 
 
 L^;^^^ 
 
 
 WILSON on CROWN-GALL 
 
BOTANICAL GAZETTE, LXX 
 
 PLATE VIII 
 
 
 WILSON on CROWN-GALL 
 
BOTANICAL GAZETTE, LXX 
 
 PLATE IX 
 
 WILSON on CROWN-GALL 
 
BOTANICAL GAZETTE, LXX 
 
 r^ .... 
 
 
 
 
 >r ■■■*t-U/^"^..J -^J ■■* oJ/r ■• 
 
 WILSON on CROWN-GALL 
 
LIBRARY OF CONGRESS 
 
 021 489 493 2