ADHESIVE BONDING PROPERTIES CE VARIOUS 
 
 METAES AS AEEECTED BY CHEMICAE AND 
 
 ANODIZING TREATMENTS OF THE SURFACES 
 
 (Part A - Additional tests on Anodized Aluminum and on 
 linr-Cliromatc-Primcd Magnesium) 
 February 1955 
 
 This Report is One cf a Scries 
 
 Issued in Cooperation with 
 
 AIR fORCE-NAVy-CIVIL SUBCOMMITTEE 
 
 en 
 
 AIRCRAFT DESION CRITERIA 
 
 Under the Supervision off the 
 
 AIRCRAFT COMMITTEE 
 
 of the 
 
 MUNITIONS BOARD 
 
 No. 1842-A 
 
 UNITED STATES DEPARTMENT OF AGRICULTURE 
 FOREST SERVICE 
 nr FOREST, PRODUCTS LABORATORY 
 
 M/adison 5, W isc on si n 
 
 MAR * r ««■ ' n Cooperation with the Univer»ity of Wisconsin 
 
 I 6 1955 
 
 nesiry of aei . 
 
ADHESIVE BONDING PROPERTIES OF VARIOUS METALS AS AFFECTED BY 
 
 CHEMICAL AND ANODIZING TREATMENTS OF THE SURFACES" 
 
 (Part A — Additional teste on Anodized Aluminum and on 
 Zinc -Chr ornate -<Primed Magnesium) 
 
 By 
 
 H. W. EICKNER, Chemical Engineer 
 
 2 
 Forest Products Laboratory, - Forest Service 
 
 U. S. Department of Agriculture 
 
 Summary 
 
 The concentration of chromic acid (5 or 10 percent) and the length of 
 the anodizing period (20 or ^0 minutes) were not found to influence the 
 adhesive bonding properties of the chromic -acid-anodi zed, bare 75S-T6 
 aluminum alloy. However, seal treatment of the anodized surface by 
 heating in water greatly interfered with bonding except for one type of 
 adhesive (high-temperature formulation of neoprene, nylon, and phenol 
 resins on fabric carrier). This same adhesive also gave unusually good 
 bonding to zinc -chr ornate -primed magnesium surfaces with bond strengths 
 averaging as high as 2,770 pounds per square inch on the standard 1/2- 
 inch overlap specimens Df 0.06if-inch- thick FS1-H magnesium. 
 
 Introduction 
 
 The following tests were made to supplement the previous work on adhesive 
 bonding of various metal surfaces reported in Forest Products Laboratory 
 
 —This progress report is one of a series prepared and distributed by the 
 
 Forest Products Laboratory under U. S. Navy Bureau of Aeronautics 
 
 Order No. NAer 01628 and U. S. Air Force Order No. AF l8(600)-102, 
 
 Amendt. No. A6( 55-286). Results here reported are preliminary and may 
 
 be revised as additional data become available. 
 
 2 
 
 "Maintained at Madison, Wis., in cooperation with the University of 
 
 Wisconsin. 
 
 Rept. No. lQk-2-A -1- Agriculture-Madison 
 
Report No. 1842.— These new tests include work to determine the effect 
 of chromic acid concentration (5 or 10 percent by weight), length of 
 treatments (20 or kO minutes), and hot-water sealing in anodizing bare 
 75S-T6 aluminum, on the initial bond strength of joints to this metal. 
 Adhesive bonding tests were also made on magnesium surfaces that had 
 been treated and zinc-chromate-primed by two aircraft fabricators using 
 their standard surface treatments. The zinc-chromate-prime coatings 
 used with these treatments were thinner than those used on magnesium 
 surfaces in the previous work in Report No. 1842, and were reported to 
 improve bonding to this metal. 
 
 Procedure 
 
 Type and Number of Specimens 
 
 Small k- by 5-l/2-inch test panels were prepared by overlapping two 3~ 
 by 4-inch metal sheets, 0,,064-inch in thickness, for l/2 inch. Metal 
 sheets of 2 types, bare 75S-T6 aluminum alloy and FS1-H24 magnesium 
 alloy, were included in these tests. Three test panels were bonded for 
 each of the variables studied with three 1-inch-wide test specimens cut 
 from each of the panels. 
 
 Preparation of Metal Surfaces 
 
 Aluminum , --The bare 75S-T6 aluminum surfaces were prepared for bonding 
 by first wiping with a cloth saturated with acetone, and then immersing 
 for 5 minutes at 170° to 190° F. in an alkaline solution (pH approxi- 
 mately 12.0) of the composition: 
 
 5.0 ounces sodium metasilicate 
 0.5 ounce Nacconal NR 
 1.0 gallon water 
 
 The alkaline solution was rinsed from the surfaces in hot water and 
 then cold water. The sheets were then anodized in a chromic acid 
 solution (5 or 10 percent by weight) at 90° to 100° F. by first apply- 
 ing a small direct current potential (3 to 5 volts) between the alumi- 
 num pieces suspended as anodes and the cathode, which was the lead con- 
 tainer for the solution. This voltage was gradually increased (within 
 5 minutes) to kO volts. The higher voltage was applied at a current 
 density of 2.5 to 3-0 amperes per square foot of anode surface for 
 periods of 20 or kO minutes. 
 
 3 
 
 — Eickner, H. W. Adhesive Bonding Properties of Various Metals as 
 
 Affected by Chemical and Anodizing Treatments of the Surfaces. 
 
 Forest Products Laboratory Report No. lQk2. 
 
 Rept. No. 18U2-A -2- 
 
After removal from the anodizing treatment, the surfaces were either 
 rinsed in warm water and force dried in front of a circulating fan 
 (unsealed), or sealed by heating the metal sheets in distilled water 
 for 1 hour at l80° F. before drying. 
 
 With the use of 2 concentrations of anodize solution, at 2 anodizing 
 periods for both sealed and unsealed surfaces, 8 types of anodized 
 aluminum surfaces were therefore prepared for investigation in this study, 
 
 Magnesium . --The surfaces of the magnesium sheets were prepared for bond- 
 ing by two aircraft fabricators by using their established procedures. 
 
 A. One method, involving electrolytic treatments, was reported to con- 
 sist essentially of the following steps: 
 
 (1) Vapor degreasing surfaces in stabilized trichloroethylene to 
 
 remove oils and contaminants 
 
 (2) Cathodic cleaning (k.O to 6.0 volts direct current at 5 to 
 
 25 amperes per square foot of cathode surface) for 3 to 10 
 minutes at 180° to 200° F. in a solution of the composition: 
 
 10.0 ounces sodium hydroxide 
 1.0 gallon water 
 
 (3) Rinse in cold water 
 
 (k) Pickling for 3 to 5 seconds at room temperature in a solution 
 of the composition: 
 
 ^•3 ounces concentrated sulfuric acid (sp. gr.-l.8J4-) 
 2.8 ounces concentrated nitric acid (sp. gr.-lA2) 
 1.0 gallon water 
 
 (5) Rinse in cold water 
 
 (6) Electrolytic treatment for approximately 15 minutes with a 
 
 k.O- to 8.0-volt, 60-cycle alternating current at a current 
 density of 20 to 30 amperes per square foot of magnesium 
 surface in a solution of the composition: 
 
 ^■0.0 ounces sodium hydroxide 
 
 0.5 ounce phenol 
 
 5.2 ounces sodium silicate (kl° Be.) 
 
 Water to make 1 gallon 
 Operating temperature of the solution was l80° to 200° F. 
 
 (7) Rinse in hot water 
 
 Rept. No. 18^2-A -3- 
 
(8) Neutralizing for 1 to 5 minutes at 135° to lk3° F. in solution 
 
 (pH-2.U) of the composition: 
 
 0.07 ounce chromic acid 
 1.0 gallon water 
 
 (9) Force drying in hot air 
 
 (10) Priming by dipping in a zinc chromate primer (MIL-P-6889A, 
 
 Type i) thinned to result in a dry film thickness of 
 0.00020 to O.OOO35 inch 
 
 (11) Air drying prime coat for at least 2k hours prior to bonding. 
 
 B. The other method, involving only chemical treatments, was reported to 
 consist essentially of the following steps: (Metal pieces with this 
 method were slightly larger, 5- by 6-inches, than the standard size.) 
 
 (1) Vapor degreasing surfaces in stabilized trichloroethylene to 
 
 remove oils and contaminants 
 
 (2) Cleaning surfaces by immersion for 10 minutes at 170° to 190° 
 
 F. in a commercial phosphoric-silicate cleaner for magnesium 
 (5 ounces per gallon) 
 
 (3) Rinse in cold water 
 
 (k) Removing scale by immersion for 1 to 10 minutes at room temp- 
 eratures in a solution of the composition: 
 
 2^.0 ounces chromic acid 
 k.O ounces sodium nitrate 
 1.0 gallon water 
 
 (5) Rinse in hot water 
 
 (6) Pickling for 5 minutes at room temperature in a solution of 
 
 the composition: 
 
 1 part by volume hydrofluoric acid ( 50 percent) 
 
 2 parts by volume water 
 
 (7) Rinse in cold water 
 
 (8) Sealing by boiling for 30 minutes in a solution of the 
 
 composition: 
 
 16.0 ounces sodium dichr ornate 
 0.2 ounce calcium fluoride 
 1.0 gallon water 
 
 Rept. No. 18^2-A - U - 
 
(9) Rinse in cold water 
 
 (10) Forced drying in hot air 
 
 (11) Priming by dipping in zinc chromate primer, thinned with 
 
 toluene (l part primer to 3 parts thinner) to a consis- 
 tency to result in a dry film thickness of 0.00015 to 
 0.00025 inch 
 
 (12) Air drying prime coat 2k hours, and then curing for 30 
 
 minutes at 180° to 200° F. 
 
 These two methods were essentially the same as methods M-5 and M-3> 
 respectively, in the original work of Report No. 18*4-2, except that the 
 prime coat in the earlier work was much thicker, and different pro- 
 prietary primers were used. 
 
 Adhesive Bonding Processes 
 
 The following k adhesive bonding processes were used in bonding the 
 lap-joint panels prepared with the 2 metals having several surface 
 conditions. 
 
 Bloomingdale FM-^T « — A high-temperature-setting formulation of the 
 vinyl-phenolic type supplied by Bloomingdale Rubber Company, Delaware 
 and Flower St., Chester, Pa. 
 
 Redux E, Type R . — A high-temperature-setting, two-component formulation 
 of a phenol-resin solution and vinyl -polymer powder supplied by Ciba 
 Company, 627 Greenwich St., New York Ik, N. Y. 
 
 Metlbond MN3C Nylon Tape . — A high-temperature-setting adhesive formula- 
 tion of neoprene, nylon, and phenol resins, supported as a film on 
 nylon-fabric tape, supplied by NARMC0 Resins and Coatings Company, 600 
 Victoria St., Costa Mesa, Calif. 
 
 Scotchweld Bonding Film AF-6. — A high-temperature-setting formulation 
 of acryonitrile-butadiene rubber and phenol resin in the form of an 
 unsupported tape supplied by Minnesota Mining and Manufacturing Company, 
 ^11 Piquette Ave., Detroit, Mich. 
 
 These adhesives are of the same general types as used in the original 
 studies of Report 18^2 and addendum, but the adhesive products included 
 are identical in only 1 of the k processes. 
 
 The conditions of bonding with each of the adhesives were as follows: 
 
 Bloomingdale FM-Vf . — Three spray coats of the adhesive, thinned with 
 1-1/2 parts by volume of adhesive solvent to 1 part of adhesive, were 
 
 Rept. No. 18^2-A -5- 
 
applied to the metal with 1-hour air drying between coats, and overnight 
 
 - drying after the final coat. The adhesive film was then precured 
 for 1 hour at 150° F. in an oven. Following the precure, the joint was 
 assembled and placed in a hot press, where it was preheated without 
 pressure for 5 to 9 minutes at 300° F. before it was given the final 
 cure for 15 to 25 minutes at 300° F. and 200 pounds of pressure per 
 square inch. 
 
 Redux E, Type R .--0ne medium coat of the liquid component was brushed 
 on the metal, and the powdered component was sprinkled immediately into 
 the wet spread of adhesive. Any excess powder was brushed from the 
 surface. The adhesive film was air dried overnight, and the joint was 
 assembled and pressed at 200 pounds per square inch of pressure for 25 
 minutes in a hot press at a temperature of 300° F. Approximately 10 
 minutes of the pressing period were required to bring the temperature 
 of the glue line to that of the press platen. 
 
 Metlbond MN3C Nylon Tape . --Four spray coats of the priming component 
 (M3C) were applied to the metal pieces to result in a 0.001- to 0.002- 
 inch film of adhesive. The adhesive film was dried for 30 minutes be- 
 tween coats and 2-l/2 to k hours after the final coat. The joint was 
 assembled with a single layer of tape adhesive and pressed at 50 pounds 
 per square inch pressure for k-2 minutes in a hot press at a temperature 
 of 335° F. Approximately 12 minutes of the pressing period were required 
 to bring the temperature of the glue line to that of the press platen. 
 
 Scotchweld Bonding Film AF-6 . — The single film of adhesive was assembled 
 in the joint. The assembly was pressed for 45 minutes at 150 pounds per 
 square inch of pressure in a hot press at a temperature of 325° F. 
 Approximately 10 minutes of the pressing period were required to bring 
 the temperature of the glue line to that of the press platen. 
 
 To insure uniform pressure distribution, all joints were pressed by 
 using cauls of 0.027-inch-thick chipboard between the press platen 
 and the metal pieces. 
 
 Testing 
 
 The 3 lap-joint panels prepared with each metal, surface treatment, and 
 bonding process were sawn into individual 1-inch-wide specimens. Cutting 
 was done with a metal-cutting bandsaw using a slow rate of feed and a 
 holding jig to minimize any mechanical damage or overheating of the 
 joint. The lap-joint specimens were tested to failure by loading them 
 in tension at a rate of 300 pounds per minute. The ends of the speci- 
 mens were held in 1-inch-wide Templin-type grips that extended down 
 from the ends of the specimens to within 1-inch of the edge of the 
 overlap. Testing was done at a temperature of 72° to 76 F. "he 
 fai] oad (calculated as unit stress on the measured test area) 
 
 Rept. No. 18^2-A -6- 
 
and estimated areas (expressed as percentage of the total area) of 
 adhesion, cohesion, and primer and coating failures were recorded. 
 
 Test Results 
 
 The results of the bonding tests made to the chromic-acid-anodized 
 aluminum, unsealed and sealed, are given in table 1, and to the zinc- 
 chr ornate -primed magnesium in table 2. 
 
 The results in table 1 definitely show that the hot-water sealing of 
 the aluminum oxide film formed during chromic-acid anodizing interferes 
 with bonding. This interference is much greater for types of adhesives 
 represented by Bloomingdale FM-kj , Redux E, and Scotchweld Bonding Film 
 AF-6 than it is for the type of adhesive represented by Metlbond MN3C 
 Tape. The concentration of the chromic-acid anodizing solution ( 5 or 
 10 percent) and the length of the anodizing period (20 or ^0 minutes) 
 were not found to influence the bonding properties of the anodized 
 surfaces significantly. 
 
 The results in table 2 again confirm the results indicated in the 
 original work and given in Forest Products Laboratory Report No. 18^2; 
 namely, that adhesives of the type of Metlbond MN3C Tape produce bonds 
 to zinc-chromate-primed magnesium sheets that usually have higher bond 
 strengths than those obtained with the Redux E and Bloomingdale FK-k'J- 
 type adhesives. The latter two types are normally judged by lap-joint 
 tests with clad aluminum to have higher bond strengths than the first 
 type. The type of solvent in an adhesive, and its reaction with the 
 primer, might be expected to influence the bond strengths to the zinc- 
 chromate-primed magnesium but this possible effect was not investigated. 
 The average shear strength of 2,770 pounds per square inch for the bonds 
 made with Metlbond Tape to the zinc-chromate-primed magnesium, treated 
 by method B, is considered by present standards to be unusually good 
 bonding to magnesium. The Metlbond Tape adhesive, and also Scotchweld 
 AF-6 film, gave bond strengths of about 1,950 pounds per square inch to 
 the magnesium treated by method A. 
 
 Rept. No. 181+2-A -7- 
 

 
 
 Cool 
 
 
 p> 
 
 
 5 
 
 
 o 
 
 
 
 ri 
 
 
 
 
 U 
 
 
 CO 
 
 
 •H 
 
 
 CO 
 
 
 o 
 
 
 Cm 
 
 
 CO 
 
 H 
 
 
 M 
 
 
 tj 
 
 
 a 
 
 •H 
 
 cd 
 
 
 cu 
 
 
 H 
 
 < 
 
 CH 
 
 
 
 
 
 i 
 
 
 
 .a 
 
 
 
 ,ri 
 
 6; 
 
 CO 
 
 
 p i 
 
 • 
 
 
 o 
 
 bo 
 
 p 
 
 bO i 
 
 •H 
 
 
 P 
 
 
 cd 
 
 a 
 
 ri i 
 
 • 
 
 
 o 
 
 
 M 
 
 •H 
 
 CO 1 
 
 CO 
 
 
 a 
 
 
 CO 
 
 o 
 
 U I 
 
 • 
 
 
 CO 
 
 
 > 
 
 "S 
 
 p 
 
 CO 1 
 
 Pm 
 
 
 
 
 Coo| 
 
 
 P 
 
 
 
 
 o 
 
 
 
 ri 
 
 
 
 
 •H 
 CO 
 
 S 
 
 
 CU 
 
 o 
 
 
 
 
 CO 
 
 . 
 
 
 u 
 
 
 •s 
 
 cu 
 
 S 
 
 •ri 
 
 cd 
 
 
 CU 
 Pm 
 
 
 9 
 
 < 
 
 <H 
 
 
 
 
 
 
 XI I 
 
 
 
 B 
 
 CO 
 
 
 •P i 
 
 • 
 
 1 
 
 nd 
 
 bo p 
 
 bO I 
 
 •M 
 
 1 
 
 cu 
 
 p 
 
 cd 
 
 
 
 C i 
 
 • 
 
 
 K 
 
 
 h 
 
 •H 
 
 CO l 
 
 CO 
 
 H 
 
 
 
 CO 
 
 O 
 
 U I 
 
 • 
 
 o 
 
 
 
 !> 
 
 •«~3 +- 
 
 Pm 
 
 
 
 
 < 
 
 
 CO 
 
 
 
 
 Cool 
 
 
 p 
 
 p 
 
 
 
 O 
 
 <D 
 
 
 ri 
 
 3 
 
 U 
 
 oo 
 
 CO 
 
 •H 
 CO 
 
 S* 
 
 
 CO 
 
 o 
 
 m 
 
 % 
 
 Ph 
 
 CO 
 
 H 
 
 
 u 
 
 CO 
 
 U 
 
 cd 
 p 
 
 € 
 
 •H 
 
 cd 
 
 
 cu 
 
 Ph 
 
 p 
 en 
 
 a 
 o 
 
 o 
 
 < 
 
 Cm 
 
 
 
 
 
 A 
 
 
 S 
 
 .a 
 
 H 
 
 CO 
 
 
 p 1 
 
 • 
 
 H 
 
 >> 
 
 bO p 
 
 W 
 
 •H 
 
 
 P> 
 
 ri 
 
 cd 
 
 a 
 
 ri 
 
 • 
 
 
 CO 
 
 
 m 
 
 •H 
 
 cu 
 
 CO 
 
 
 s 
 
 
 CO 
 
 o 
 
 u 
 
 • 
 
 
 
 
 > 
 
 "O +- 
 
 Pm 
 
 
 
 
 < 
 
 
 CO 
 
 
 
 
 
 riool 
 
 O CO 
 
 
 P 
 
 ri 
 
 
 cu 
 
 
 •H 
 CO 
 
 8 
 
 
 CO 
 
 o 
 
 
 H 
 
 
 CO 
 
 H 
 
 
 M 
 
 
 3 
 
 t— 
 
 3 
 
 •H 
 cd 
 
 
 CU 
 Ph 
 
 
 bO-4" 
 
 -a! 
 
 cm 
 
 
 
 
 M 
 •H 
 
 ■ 
 
 s 
 
 
 
 ;d 
 
 
 
 1 
 
 Pn 
 
 CO 
 bo P 
 
 § 
 
 •H 
 
 
 o 
 
 
 cd 
 
 a 
 
 ri 
 
 • 
 
 
 H 
 
 
 fn 
 
 •H 
 
 CO 
 
 CO 
 
 
 PQ 
 
 
 CO 
 
 o 
 
 U 
 
 • 
 
 
 
 
 > 
 
 "~3 P 
 
 Ph 
 
 
 
 
 <J 
 
 
 CO 
 
 
 
 
 p> 
 
 
 
 
 
 
 
 a 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 H 
 
 a 
 
 
 
 
 
 
 cd 
 
 p 
 
 
 
 
 
 P 
 
 cu 
 
 cd 
 
 
 
 
 
 a 
 
 CO 
 
 CO 
 
 
 
 
 
 cu 
 
 
 h 
 
 
 
 
 
 
 
 p 
 
 cd 
 
 
 p 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 a» 
 
 SI 
 
 tj 
 
 
 
 
 ■ 
 
 m 
 
 •H 
 
 o 
 
 
 
 
 a 
 
 p 
 
 O 
 
 ■rl 
 
 h 
 
 
 
 
 •H 
 
 2 
 
 cu 
 3 
 
 4 
 
 CO 
 P< 
 
 
 
 
 
 i 
 
 
 
 
 
 
 o 
 
 cd 
 
 
 
 
 
 p 
 
 5 
 
 m 
 
 CH c 
 
 j 
 
 
 
 ri 
 
 P 
 
 ri 
 
 O i 
 
 * 
 
 
 
 cu 
 o 
 
 
 CO 
 
 ri ' 
 
 3 -H 
 
 
 
 u 
 
 
 o 
 
 O ( 
 
 H O 
 
 
 
 cu 
 
 
 ri 
 
 •H .1 
 
 3 cd 
 
 
 
 Ph 
 
 
 o 
 
 p < 
 
 j 
 
 
 
 
 
 c_> 
 
 
 
 
 
 
 o oo 
 
 O H 
 H 
 
 OONO 
 H OJ 
 O OO 
 •\ »\ 
 H -4- 
 
 -d- CO 
 NO H 
 
 NO 
 
 -d 
 t— 
 
 -4" 
 
 OJ OO 
 
 O On 
 O OD 
 
 OOVO 
 
 CO o 
 
 o\ o 
 on 
 
 NO ON 
 [■— OO 
 
 H NO 
 C-- OJ 
 
 a\ oo 
 
 ON H 
 
 OJ OO 
 
 o co 
 
 O CO 
 H 
 
 ON OO 
 
 NO -3- 
 CO -H- 
 
 bO=H 
 
 ri m 
 
 •H CU 
 
 H P 
 
 ■g g 
 
 pq 
 
 b0-4-| 
 
 ri Jh 
 
 •H 
 CU H 
 
 ri -h 
 
 O O 
 
 £5 pq 
 
 u 
 
 p 
 
 a 
 
 ctn 
 
 O -d 
 
 CO OO 
 
 OJ no 
 
 H OO 
 
 O LT\ 
 
 O OJ 
 H 
 
 NO NO 
 OJ CM 
 H -d 
 
 c— CO 
 t— OJ 
 
 -d- ON 
 vO ON 
 
 OJ OJ 
 
 o 
 o 
 
 
 O OO 
 
 o -=«■ 
 
 ON t— 
 
 On ltn 
 CO t*- 
 
 »\ 
 OO 
 
 o o 
 
 O H 
 
 CO H 
 
 go r— 
 
 NO OJ 
 
 co oj 
 
 OJ H 
 
 ^" LTN 
 l/N OJ 
 
 OJ OO 
 
 O H 
 O ON 
 
 H 
 
 LTN OJ J" OO 
 
 LT\CO LTNCO 
 
 NO H LTN^J- 
 
 ri^L 
 
 •H CO 
 H P> 
 
 •s g 
 
 pq 
 
 ri^M 
 
 CO H 
 
 ri -h 
 
 O o 
 
 s pq 
 
 cu 
 p 
 
 8 
 
 5 
 
 s 
 
 ^ 
 
 •^ 
 
 K 
 
 CO 
 
 > 
 
 •H 
 CO 
 
 cu 
 
 cd 
 cu 
 p 
 
 a 
 
 CO 
 
 % 
 
 p 
 
 CO 
 £> 
 
 c 
 o 
 
 •H 
 CO 
 CO 
 
 cu 
 
 8 
 
 H 
 
 •H 
 
 cd 
 cm 
 
 ^1 
 
 o 
 
 cd 
 cu 
 
 cd 
 
 cu 
 
 c 
 o 
 
 Cm 
 
 o 
 
 cu 
 xi 
 p> 
 
 Cm 
 
 o 
 
 cu 
 
 p 
 
 cu 
 
 CU CO 
 
 •n x: 
 
 •M P 
 
 ■H ftft 
 
 -. t: 
 
 p CO 
 
 cd T3 
 
 p 
 
 B cu 
 
 •H t^ 
 
 c 
 cu 
 
 P -H 
 
 B 
 
 co tJ 
 
 p 
 
 CO o 
 
 cd 
 
 c 
 
 cu 
 
 co cd 
 
 M 
 
 bO 
 
 p 
 
 cd TJ 
 
 
 m ri 
 
 H 
 
 cu cd 
 
 cd 
 
 > 
 
 CO 
 
 < 
 
 CO 
 
 oo| 
 
 -*l 
 
 CO t3 
 
 -a 
 
 Kept. No. 18U-2-A 
 
 -8- 
 
u 
 
 Pi 
 
 I 
 
 CD 
 
 I 
 
 o 
 I 
 
 CJ 
 
 d 
 
 •H 
 CO 
 
 CH 
 
 o 
 bD 
 
 a 
 
 •H 
 
 •d 
 d 
 o 
 
 CD 
 > 
 •H 
 
 ra 
 
 OJ 
 
 (L) 
 H 
 X) 
 
 CM 
 
 
 
 
 
 bO 
 
 p 
 
 H 
 
 
 
 u 
 
 
 d 
 
 i d 
 
 cd 
 
 
 
 53 
 
 
 •H 
 
 1 CD 
 
 0) 
 
 
 
 a 
 
 Jh 
 
 -P 
 
 1 o 
 
 m 
 
 
 
 •rH 
 
 
 
 O 
 
 1 CD 
 
 xJ 
 
 
 
 PM 
 
 
 O 
 
 PH 
 
 s 
 
 i-j 
 
 
 
 
 
 
 
 
 
 
 p 
 
 d 
 
 : a 
 
 
 d 
 
 O 
 
 
 u 
 
 P 
 
 53 
 
 i H 
 
 
 •H 
 
 
 53 
 
 d 
 
 a 
 
 1 -H 
 
 
 CO 
 
 o 
 
 3 
 
 CD 
 
 -p 
 
 i cd 
 
 
 CD 
 
 -p 
 
 CJ 
 
 cd 
 
 CH 
 
 
 3 
 
 
 rH 
 
 fn 
 
 a) 
 
 
 
 
 Ph 
 
 CD 
 
 h 
 
 <P 
 
 
 <U 
 
 
 
 Ph 
 
 -p 
 
 o 
 
 CD 
 
 
 
 
 
 
 H 
 
 
 
 
 
 03 
 
 ft 
 
 
 d 
 
 
 
 
 cj 
 
 b>j 
 
 
 o 
 
 
 
 P 
 
 fi 
 
 ^ 
 
 
 •H 
 CO 
 
 
 
 d 
 
 CD 
 
 CD 
 
 
 
 CD 
 
 
 
 CJ 
 
 XI 
 
 
 
 xl 
 
 
 
 U 
 
 o 
 1 
 
 
 
 O 
 
 o 
 
 
 
 CD 
 Ph 
 
 A 
 
 
 
 X) 
 
 
 
 
 -d 
 
 CD 
 
 
 -p 
 
 
 
 • 
 
 o 
 
 bO P 
 
 bO 
 
 
 
 •H 
 
 XJ 
 
 cd 
 
 ^ 
 
 d 
 
 
 
 • 
 
 -p 
 
 rH 
 
 •H 
 
 CD 
 
 
 
 CQ 
 
 .22 
 
 CD 
 
 O 
 
 rH 
 
 
 
 • 
 
 a 
 
 > 
 
 •r-3 + 
 
 
 
 Ph 
 
 
 < 
 
 
 CO 
 
 
 
 
 
 
 
 
 
 bD 
 
 -P 
 
 
 
 
 rH 
 
 
 d 
 
 d 
 
 
 
 
 CD 
 
 
 •H 
 
 CD 
 
 h j. 
 
 
 
 R 
 
 rH 
 
 -P 
 
 CJ 
 
 
 
 •H 
 
 o 
 
 cd 
 
 rH 
 
 d 
 
 
 
 Jh 
 
 
 o 
 
 CD 
 
 53 
 
 
 
 Ph 
 
 
 CJ 
 
 Ph 
 
 g 
 
 p 
 cd 
 
 CO 
 
 1 
 
 
 
 
 
 
 
 d 
 
 
 
 
 rH 
 
 H 
 
 
 O 
 
 
 U 
 
 P 
 
 -p 
 
 •H 
 
 
 •H 
 
 
 53 
 
 d 
 
 
 cd 
 
 
 CQ 
 
 o 
 
 1 
 
 CD 
 
 o 
 
 CH 
 
 
 53 
 
 -p 
 
 CJ 
 
 •H 
 
 
 
 3 
 
 
 u 
 
 H 
 
 £ 
 
 «H 
 
 
 
 ft 
 
 CD 
 
 O 
 
 
 <JJ 
 
 
 
 Ph 
 
 H 
 
 O 
 
 U 
 
 -P 
 
 0) 
 
 g 
 
 
 
 
 
 
 
 d 
 
 
 
 
 o 
 
 EH 
 
 
 o 
 
 
 
 -P 
 
 0) 
 
 
 
 •H 
 
 
 
 d 
 
 H 
 
 
 
 CO 
 
 
 
 CD 
 
 i 
 
 
 
 CD 
 
 Xl 
 
 o 
 u 
 
 
 
 O 
 
 rH 
 
 CD 
 Ph 
 
 < 
 
 
 
 
 
 
 
 ^d 
 
 
 
 XJ 
 
 
 
 
 o 
 
 X) 
 
 CD 
 bO -P 
 
 ft 
 
 
 
 • 
 
 •H 
 
 -p 
 
 cd 
 
 q 
 
 d 
 
 
 
 • 
 
 .SJ 
 
 fn 
 
 •H 
 
 CD 
 
 
 
 CQ 
 
 S 
 
 
 
 o 
 
 rH 
 
 
 
 • 
 
 
 t> 
 
 *«~3 -P 
 
 
 
 Ph 
 
 
 < 
 
 
 CQ 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 > 
 
 
 
 
 
 
 
 ■H 
 
 
 
 
 
 
 
 CO 
 
 
 
 
 
 
 
 CD 
 
 
 
 
 
 
 
 € 
 
 
 
 
 
 
 
 < 
 
 
 
 
 
 
 
 o 
 o 
 
 L^ 
 
 CO 
 
 OJ 
 
 UA 
 
 LTN 
 
 -4" 
 
 -4 
 
 O 
 
 o 
 
 o 
 
 LT\ 
 
 r- 
 
 t*- 
 
 on 
 
 o 
 
 t- 
 
 co 
 
 o 
 
 LT\ 
 
 o 
 
 o 
 
 CO 
 
 o 
 
 H 
 
 
 H 
 
 cr\ 
 
 -d- 
 
 t- 
 
 vo 
 
 on 
 
 vo 
 
 t- 
 
 ON 
 
 H 
 
 o 
 o 
 
 o 
 
 o 
 
 CO 
 
 on 
 vo 
 
 ON 
 
 on 
 
 CO 
 H 
 
 on 
 
 o cd 
 
 CD H 
 bD 
 cd cd 
 
 t3 
 
 CD xl 
 
 bO -H 
 
 d -n 
 
 4H CD 
 
 cd 
 
 xi cd o a 
 
 cd cq ir\ M 
 
 <+H CD 
 
 -P CD 
 cd-£ £ 
 
 Ph 
 
 > en 
 
 -p <h 
 cq o 
 
 cd cd 
 
 <+H o 
 
 CD 
 
 ^ bD 
 
 Rept. No. 18^2-A 
 
 -9-