J u n e , 1911 storms.
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING CHEMISTRY
Such a comparison is given in Table 111.
459
have been proposed a n d thoroughly tested, b u t as they on certain derivatives of rubber, t h e nitrosite, t h e nitrosate, a n d t h e tetrabromide, t h a t have not t h u s far been obtained with unvarying composition, these methods have not found general acceptance either as technical or research methods for this extremely important estimation. It is hoped t h a t t h e method' about t o be described will lend itself t o development not only as a dependable one for rubber research work, b u t also for t h e commercial laboratory. The procedure, in brief, consists first in forming the nitrosite of rubber b y t h e action of nitrogen trioxide gas upon a finely ground a n d acetone-extracted sample of t h e rubber suspended in chloroform. After t h e completion of t h e action, t h e insoluble nitrosite, fillers, etc., are filtered from t h e chloroform, a n d t h e nitrosite is dissolved in acetone. T h e suspension of finely divided mineral matter is t h e n allowed t o settle o u t , or is thrown down with t h e centrifuge. An aliquot portion of t h e solution is transferred with a pipette t o a small flask, a n d its volume reduced b y evaporation t o a few cubic centimeters. This small volume of acetone solution of t h e nitrosite is now transferred with t h e help of ethyl acetate t o a porcelain boat containing alundum, and, after t h e acetone a n d ethyl acetate have been expelled b y warming t h e boat for several hours in a drying oven, t h e nitrosite is burned in a current of oxygen, a n d t h e carbon dioxide t h u s formed is absorbed in soda-lime a n d weighed. If all of t h e carbon originally in t h e sample as rubber, a n d only such carbon, reaches t h e soda-lime apparatus as carbon dioxide through t h e intermediate nitrosite, t h e equation C10H16 + IoCOZ
TABLEIII(a)--cOMPARISON OF DATAO N EUROPEAN SIROCCO DUST FALLS depend for their accuracy WITH THAT FOUND NEAR CEMENTPLANTS
DATE Oct. 16, 1846 M a r c h 31, 1847 1859 Feb., 1862 March 24, 1869 March 19, 1901 March 9-12, 1901
PLACE Southeastern France Tyrol Westphalia Salzburg, Austria Carniola, Austria Taormina, Sicily EuroDe
1913 ( b y author) 1912 ( b y author)
California. California.
Weight of d u s t in pounds per acre per d a y 5.62 17.80 275.0 0.75 44.7 24.1 32.4 3.02 10.9
Maximum MiniGiG Maximum in Table I Maximum found near a cement plant 22.9 1913 (by author) California. Lowest d u s t fall f a r f r o m a cement plant 0.42 (a) T h e d a t a used for comparison in this table were obtained from Bull. 68, Bureau of Soils, United S t a t e s Department of Agriculture, a n d have been recalculated into terms of pounds per acre per d a y . Where only the d a t e of t h e d u s t fall was given the time was assumed t o b e one day.
T h e question of t h e elimination of t h e loss of cement dust by t h e installation of proper devices a t t h e plant does not come within t h e scope of this paper. It is sufficient t o s t a t e t h a t i t can be accomplished, a n d t h a t it is a m a t t e r of record t h a t amounts in excess of fifty tons per d a y of cement dust have been recovered from a five-kiln plant in such a form a s t o permit its use in t h e manufacture of cement, a n d t h a t t h e result has been advantageous t o both t h e cement company a n d its neighbors. CONCLUSION
I n this paper t h e problem arising from t h e loss of large amounts of dust through t h e stacks of cement plants has been outlined. T h e methods available for t h e determination of t h e dust fall in t h e field have been stated a n d discussed. A practical method has . been given in detail, a n d 2 method of calculation b y which t h e cement dust can be distinguished from t h e field dust has been explained. T h e importance of a s t u d y of t h e wind a n d weather conditions a s a preliminary t o proper interpretation of t h e results of field work has been emphasized. D a t a have been presented which show the results obtainable by t h e methods given, a n d t h e relation of t h e dust fall near cement plants t o t h a t of certain great d u s t storms. The writer wishes t o express his indebtedness t o Mr. G. S. Bohart, of this University, for his assistance in carrying out t h e analytical p a r t of t h e work which has been presented. DEPARTMENT O F CHEMISTRY
STANFORD UNIVERSITY,CALIFORNIA
THE COMBUSTION METHOD FOR THE DIRECT DETERMINATION OF RUBBER' B y I,. G. W s s s o ~ Received M a r c h 9, 1914
T h e method most in use a t t h e present time for t h e determination of t h e caoutchouc content of rubber goods is an indirect one, in which t h e sample is analyzed for its moisture, mineral m a t t e r , sulfur, resin a n d other contents, these values being t h e n added together a n d t h e s u m subtracted from IOO per cent t o arrive a t t h e percentage of rubber present. As many of t h e best methods for estimating these constituents are admittedly inaccurate, t h e indirect method is not a satisfactory one. A number of direct methods 1
Published b y permission of t h e Director of t h e Bureau of Standards.
enables one t o calculate t h e C I O H Ior~ real caoutchouc content of t h e sample. P R O C E D U R E A N D A P P A R A T U S I N DETAIL-The procedure a n d apparatus employed in obtaining t h e results given later are t h e outgrowth of many trials a n d experiments. Doubtless deviations are allowable in many points, b u t there was not opportunity t o s t u d y t h e effect of changing various factors. T h e sample should be ground t o pass a ao-mesh sieve, if possible, or cut up fine with t h e scissors if very soft. A weighed amount of t h e sample, 0 . 5 gram for compounds containing a b o u t 50 per cent or less of rubber, a n d 0 . 2 5 gram for those containing a higher percentage, is wrapped into a bundle with a g cm. filter paper a n d extracted for 3 t o 4 hours with acetone in a n apparatus of t h e Wiley or Cottle type, in which t h e sample is extracted by t h e solvent a t t h e boiling point of t h e latter. T h e residue, from which t h e excess of acetone has been squeezed with t h e fingers, is t h e n transferred t o 2 50 cc. calibrated, glassstoppered flask a n d allowed t o dissolve or swell up in a b o u t 40 cc. of chloroform,z which action may be hastened b y warming t h e flask. 1
A preliminary note on this method was published i n THISJOURNAL,
6 (1913), 398. 2 T h e chloroform should not be previously dried, as moisture is a p parently advantageous in giving a more rapid action of t h e nitrogen oxides on t h e rubber.
460
T H E J O U R N A L OF I N D U S T R I A L A N D ENGINEERING C H E M I S T R Y
T h e rubber is now submitted t o t h e action of nitrogen trioxide gas, evolved b.y running nitric acid of sp. gr. I . 3 dropwise on arsenic trioxide contained in a flask warmed in a boiling water b a t h . After t h e gases have been passed through a n e m p t y gas-washing bottle t o condense most of t h e moisture a n d nitric
Vol. 6, N o . 6
stoppering, for a half hour t o ensure complete solution of t h e nitrosite in t h e acetone. T h e volume is now made exactly t o t h e mark, t h e flask stoppered a n d shaken, a n d , t o obtain a clear solution, t h e insoluble mineral m a t t e r is allowed t o settle, or better, is quickly centrifuged o u t : I j o o revolutions per minute
APPARATUS FOR THE DIRECT DETERMINATION OB RUEEER BY THE COMBUSTION METHOD A. Oxygen enters 0. U-tube containing 30-mesh granulated zinc G. Platinum spiral a b o u t 10 cm. long (No. 20 wire) B. Moist soda lime H. Porcelain stem P, P. Soda lime C. Bubble counter containing sulfuric acid I. Platinum leg for supporting stem Q. Calcium chloride (Dennstedt) R. Alumina K. Platinum leads 8 cm. long (No. 10 wire) S. U-tube containing palladium chloride soluD. Closed tube of Jena combustion glass, 8 cm. L. Short platinum wires sealed into tube long. t o prevent backward diffusion M . Mercury contacts tion E. Porcelain boat N, N. U-tubes containing glass beads and HBOaT. Rheostat F. Heating coil of nichrome ribbon KzCnO7
acid carried over from t h e generator, t h e y enter t h e chloroform through a delivery t u b e joined b y a rubber connection t o t h e gas washing bottle, a n d fitting closely end for end t o t h e same. The flask containing t h e chloroform should be immersed in a beaker of cold water during t h e reaction, since t h e solubility of t h e oxides of nitrogen in chloroform is increased a n d t h e danger of t h e gas exerting a partial oxidation of t h e rubber t o COZ is probably diminished thereby. T h e gas should be passed into t h e chloroform until a deep green color is obtained which is permanent for a t least 1 5 t o 2 0 minutes after t h e delivery t u b e has been disconnected from t h e gas generator. T h e next morning t h e chloroform is decanted off, using gentle suction, through a small Gooch crucible provided with a m a t of dry asbestos. If t h e filtrate is colored brown f r o m t h e dissolved gases, one can be certain t h a t a sufficient excess of t h e nitrogen oxide has been used. After t h e flask has been rinsed out several times with small volumes ( 5 cc.) of chloroform, each time decanting carefully through t h e crucible, t h e excess of chloroform a n d acid vapors which remain in t h e flask should be expelled by t h e passage for a few minutes of a gentle air current through t h e original delivery tube. If some nitrosite from t h e rubber connection is on t h e upper end of this t u b e , i t may be easily removed b y moistening it with ac&one a n d wiping clean with a piece of filter paper. T h e nitrosite in t h e crucible is t h e n dissolved by placing t h e crucible in a 7 j cc. beaker, adding successive small portions of acetone, a n d pouring each into t h e original flask until about 40 cc. have been used. N o h a r m is done if a portion of t h e asbestos gets i n t o t h e flask. I n t h e meantime, t h e delivery t u b e has been freed from t h e nitrosite b y sucking some of t h e acetone u p into i t from t h e beaker, and. rinsing t h e outside into t h e same. Let t h e flask now remain in a beaker of water with occasional shaking, without
are sufficient t o give a clear solution in from 5 t o I O minutes. T o regain t h e original temperature, t h e flask is now allowed t o s t a n d again in t h e beaker of water until t h e correct volume is once more attained. An aliquot portion of t h e solution ( 2 j cc.) is now pipetted t o a 50 cc. Erlenmeyer flask. T o reduce t h e bulk of this solution t o a few cubic centimeters, t h e flask is warmed in a dish of water while a current of air is blown into t h e flask. T h e acetone should not be completely expelled. T h e nitrosite is transferred from t h e Erlenmeyer flask t o a porcelain boat 14 cm. long X I . I cm. wide, a n d about two-thirds full of alundum,l which is used t o secure a n even combustion of t h e nitrosite. T h e acetone solution of t h e nitrosite should not be poured into t h e boat, b u t should be drawn up i n t o a small ( 2 cc.) pipette a n d r u n out evenly over t h e alundum. Several z t o 3 cc. portions of ethyl acetate2 are t h e n used t o rinse out t h e remainder of t h e nitrosite, using a small wash bottle a n d t h e pipette for this purpose. E t h y l acetate is used for t h e expulsion from t h e nitrosite of acetone which is otherwise retained in small amounts, perhaps mechanically by t h e nitrosite or by t h e portion of t h e mineral matter which passes in solution with t h e nitrosite into t h e boat, or b y reaction with t h e nitrosite during drying. When enough wash liquid has been added t o t h e boat t o show above t h e alundum, t h e most of it should be expelled b y placing t h e boat for a few minutes in t h e drying oven. T h e same procedure should be repeated a t least once, using small portions of ethyl acetate in t h e transfer as described above. When one is certain t h a t no nitrosite remains in t h e flask or pipette, t h e boat is dried for z hours a t about 8 j C. 1 “ R R ” alundum, 90-mesh, specially prepared for carbon determinations. Norton Company, Worcester, Mass. A fresh portion should be used for each combustion. 2 T h e so-called “absolute ethyl acetate” containing about 2 per cent of alcohol. It should be redistilled before use since i t may contain other organic substances not easy t o volatilize.
,
June. 1914 ’
T H E J O I;R S A L 0 F I S D C S T RI A L A N D E SGI A\- E E RI .V G C H E M I S 1 R Y
A circulation of air in the oven t o carry a\i-ay the acetone and ethyl acetate vapors xyill assist the drying greatly. The nitrosite is now ready for the combustion. T h e combustion apparatus as i t was finally developed. contains features already used outside of this laboratory,‘ and others original with either my colleagues or myself. The arrangement of the parts is best understood from t h e figure. T h e tube is of Jena combustion glass. or bette; of quartz, j o cm. long and I . j cm. bore. The nitrosite is decomposed b y the heat of a n external coil, made by winding two layers of nichrome ribbon,? leaving 0 . I cm. between t h e t u r n s , on a n asbestos-covered copper or brass tube, j em. long and of such a diameter a s t o leave a 0 . 3 cm. space between the metal and the combustion t u b e ; and then corering it with asbestos paper t o form a n insulating layer I cm. thick. During t h e combustion the coil as it moves forward should never be moved so fast t h a t i t s forward end reaches more t h a n I cm. beyond t h e border between the black carbon of t h e undecomposed nitrosite and the white alundum of t h e completely burned portion. The decomposition products are carried forward b y t h e current of oxygen over a red-hot spiral of platinum wire which serves as t h e catalyzer for complete oxidation. The spiral is made from I . 3 meters of No. 2 0 wire wound into a cylinder 0 . 7 cm. in diameter, and is supported in the middle of the tube by a n unglazed porcelain or clay stem 1 2 cm. long and o . 3 cm. in diameter, t h e return end of the spiral passing back through t h e stem. T h e leads for the spiral are 8 cm. long and are made with S o . I O platinum wire. One end of t h e pipe-stem is supported by these leads, a n d t h e other end b y a small leg of platinum wire. T h e leads m a y connect with the outside in one of several ways. There m a y be a rubber stopper a t the forward end of the t u b e through which are pushed two heavy copper, nickel, or platinum wires, the ends of which are bent into small loops into which the platinum leads are placed before the insertion of the stopper. Sickel is the better of t h e first two metals named. T h e platinum wires are good! a s they may be seale’d into small glass tubes which reach just through the stopper. I n a n y case t h e stopper soon deteriorates, although this action has no noticeable effect on the results of the combustion. T h e most satisfactory method, however, consists in drawing down the forward end of the combustion tube as in the diagram, joining on a 3 cm. tube of 0 .5 cm. bore, a n d sealing in t w o 3 cm. lengths of heavy platinum wire. These are bent u p on t h e inside and the platinum leads are slid forward so as t o rest upon them. T h e oxides of nitrogen formed during the combustion of t h e nitrosite are absorbed b y a saturated solution of potassium bichromate in concentrated sulfuric acid. Acid vapors and sulfur trioxide are held back b y 30-mesh granulated zinc. Attention is called 1 T h e most important of these are t h e electrically heated platinum coil used as a catalyzer [Morse and Taylor, Am. Chem. J . . 33 (1905). 5911; a n d t h e electrically heated external coil used for t h e decomposition of t h e substance t o be burned. 2 0.11 X 1/32’’, R = 1.3 t o 1.5 ohms per f t . , Driver Harris Wire Co.. Harrison. N. J.
461
t o the form of apparatus in the figure for the absoi-ptjon of carbon dioxide by soda-lime. The capillary tube makes a good substitute for n stopcock on account of its comparative lightness. and is effective in separating the moist soda-lime from the c:ilcjum chloride. E m p t y , the apparatus weighs 2 0 t o 2 5 grams and will hold 3 j grams of soda-lirnc,’ and j grams of calcium chloride. I t Trill absorb I O grams or more of carbon dioxide without reiicxal. T h e second soda-lime tube contains alumina2 in its second arm, which, i t is thought, dries t h e gas t o a degree comparable. for the purposes of the present work, t o t h e drying by the concentrated sulfuric acid which precedes t h e soda-lime apparatus, T o he certain t h a t complete combustion is obtained. the gas finally passes through a faintiy yellow solution of palladium chloride in water. While using the apparatus in its present form, i t has not been necessary t o I-enew thc. pallz c1’ium chloride solution, n o sign of reduction having appeared in a number of combustions. The sulfuric acid-bichromate solution must be renewed in the first U-tube after every two or three analyses. while the zinc will last for a larger number of determinations. CALCCLATIOs--If 0 . j gram sample, and 2 j rc. out of jo cc. of t h e acetone solution h a r e been taken. the weight of C O , found multiplied b y 1.36,’440 X 4 X I O O = 1 2 3 . 6 gil-es t h e percentage of C10H16 on t h e . basis of CI0Hl6-+- roCOz. TIME REQUIRENESTS-FOI a single nitration. :LllOLlt I 5 minutes per sample are necessary; for the combustion. 30 t o 4j minutes. Analyses of samples ground up in t h e forenoon of one d a y are completed t h e next. With one combustion tube, two malyses may be run per d a y along with other work, while b y using two tubes i t is thought t h a t , after some lxoficiency has been acquired, as many as four determinations can be made per day. FORMATIOS
OF
co?
FROM
THl:
KUBBBR
DURING
is claimed by A l e ~ a n d e r .t~h a t during t h e formation of t h e nitrosite and nitrosate of rubber, large quantities of carbon dioxide are evolved from the oxidation of the rubber by t h e nitrogen oxides. G ~ t t l o b on , ~ the other hand, found only very small amounts of carbon dioxide. The writer made severaI tests of this important point by passing the gases from t h e rubber solution into a’large volume of clear, saturated baryta water. Fine Para rubber, previously extracted with acetone and dried, was used for the experiments. After the chloroform, cooled as in ?he analytical procedure, had attained the deep green color, the apparatus wa,s allowed t o stand three t o four hours, after which a current of carbon dioxitlefree air was used t o sweep into the barium hydroxide solution any carbon dioxide which might have formed during the interval. Only a trace of barium carbona t e , a slight ring in the delivery t u b e , was formed, although the barium hydroxide solution remained alkaline throughout the experiment.
NITRATIOX-It
1 T h e J. T. Baker Chemical Co. furnishes a 12-mesh soda-lime containing 15 per cent water, prepared for carbon dioxide absorption. * F . M . C. Johnson, J . A m . Chem. SOL.,34 (1912), 911. * Z . angeii’. C h e m . , 20 (190i),1358; 24 (1911). 684. 4 I b i d , 20 (19071, 2213
T H E J O U R N A L OF I N D U S T R I A L A N D E N G I N E E R I S G C H E M I S T R Y
46 2
DETERDIINATIOS
OF
SULFUR
THE
OF
VULCAXIZA-
TIoN-If t h e statement of Alexander' proves t o be true, t h a t t h e sulfur of vulcanization of t h e rubber remains quantitatively i n t h e nitrosite, this method could possibly admit of t h e simultaneous determination of t h e sulfur of vulcanization. An aliquot portion of t h e clear acetone solution of t h e nitrosite would be eyaporated t o dryness, a n d t h e sulfur determined in t h e usual way. RESULTS-The following results were obtained b y t h e method above described. All are given t h a t have been obtained on good quality, soft-vulcanized compounds since t h e d a t e after which n o great changes in t h e procedure were made. Whether t h e method is applicable t o compounds of poor quality has not been determined, a n d as t h e author is no longer in a position t o work on this point t h e field must be left t o others. This Bureau may be able, however, t o work in this field a t some later date. A washed a n d dried Up-river Fine gave 94.0, 95.1, 95.8 and 95.9 per cent CmHm Average = 95.2 per cent plus 3.3 per cent acetone extract = 98.5 per cent. B. 4 commercial compound containing 45 per cent Fine P a r a gave 42.0, 42.2, 42.i, 43.1, 43.3 a n d 43.4 per cent C I O H M Average = 42.8 per cent plus 1.3 per cent acetone extract = 44.1 per cent. C . A commercial compound containing 48 per cent Fine P a r a gave 44.8, 45.1, 45.1 a n d 45.3 per cent CIPHIG. Average = 45.1 per cent plus 2.4 per cent acetone extract = 47.5 per cent. D . T h e same compound after standing finely ground for a month gave 43 , 4 , 43.5, 43.5 and 45.0 per cent CmHm Average = 43.7 per cent. A commercial compound containing 25 per cent Fine P a r a a n d 20 per E cent Caucho or 45 per cent gum gave 40.3, 40.4, 40.6 a n d 41.1 per cent C1OH16. Average = 40.6 per cent plus 3.2 per cent acetone extract = 43.8 per cent. F . A commercial compound containing 41.5 per cent Coarse P a r a gave 39.5, 39.6, 39.8 a n d 40.1 per cent CioHie. -4verage = 39.8 per cent plus 2.1 per cent acetone extract = 41.9 per cent
A.
S U lf M A R Y
A new method for t h e direct. determination of rubber is described, which is based. upon t h e combustion of t h e nitrosite of rubber in a current of oxygen, a n d weighing of t h e carbon dioxide t h u s formed. T h e results indicate a fair degree of reliability for both raw rubber a n d high-grade vulcanized compounds. The use of this method for t h e analysis of low-grade c o m p o u n d s a n d for t h e simultaneous determination of sulfur of vulcanization m a y be possible if its applica$ion t o these fields is further studied. During this work m a n y ' valuable suggestions were made b y Dr. W. F. Hillebrand, Mr. J . B . T u t t l e , a n d a number of others a t this Bureau, a n d I t a k e this opportunity t o express t o t h e m m y appreciation of t h e same. BITREAUO F S T A N D A R D S WASHINGTON
a
OSAGE ORANGE-ITS VALUE A S A COMMERCIAL DYESTUFF* By F. W. KRESSMANS
INTRODUCTIOIi
This s t u d y is the result of a n investigation on t h e utilization of Osage orange mill waste. T h e t r u n k of t h e Osage orange tree is rather small in size, misshapen, a n d generally defective as a saw 1 2
1914.
Z . angew,. Chem.,ZO (1907). 1364; 24 (1911), 6 8 7 ; Ber., 40 (190:). 1 0 7 7 . Presented a t the 49th meeting of the A. C. S.. Cincinnati, April 6-10,
V O ~6. , NO. 6
log; a n d , although because of t h e valuable properties of t h e wood (for wagon felloes especially) closer utilization will scarcely be found in t h e use of a n y other wood, comparatively large amounts of waste are produced annually. Osage orange has long been used in Texas in a small way as a dyewood. T h e roots, b a r k , and wood are chipped a n d boiled with water a n d a more or less permanent yellow is obtained from t h e extract. Sargentl mentions t h e root bark as a source of a yellow dye a n d i t has even been suggested b y some2 t h a t Osage orange is superior t o fustic in its dyeing qualities, although no actual comparative experiments between fustic a n d Osage orange seem t o have been recorded. I n view of these facts, i t seemed advisable not only t o determine t h e chemical nature of t h e dyestuff b u t also i t s dyeing value as compared with t h e commercial dyewood it resembles most, namely, fustic. A qualitative s t u d y of t h e aqueous extract obtained from t h e wood showed t h a t t h e dyeing principles present were, as in fustic, morin or moric acid, a n d morintannic acid or maclurin. From a preliminary series of dyeing experiments made a t this laboratory, i t was found t h a t Osage orange, like fustic, is a polygenetic mordant dyestuff. Since t h e wood seemed t o contain a sufficient amount of dye t o give i t commercial value, a series of comparative dyeing experiments on fustic a n d Osage orange were arranged so as t o determine as accurately as possible t h e value of Osage orange in terms of a wellknown standard such as fustic. I n order t o have these experiments performed a t institutions best equipped for t h e purposes a n d also t o obtain t h e results of a number of different workers in this field, t h e cooperation of a number of t h e leading 'textile schools3 of t h e country was sought a n d t h e writer wishes t o take this opportunity t o t h a n k t h e m for their cooperation a n d assistance. COMPARISOX OF DEPTH O F COLORS PRODUCED
EeiLIt was noticed t h a t dyeings produced with Osage orange were weaker t h a n those obtained with fustic under t h e same conditions. Since t h e Osage orange produced a shade of color slightly differe n t from t h e fustic in most cases, i t is difficult t o determine t h e exact relative strength of t h e two products. It seems, however, t h a t , t h e Osage orange contains something like 2-25 per cent less coloring matter t h a n t h e fustic." R E P O R T B-'' Dyeings were made under identical conditions on chrome mordanted worsted yarn with t h e same amounts of both t h e wood a n d solid extracts of both substances (;. e., fustic a n d Osage orange). " T h e Osage orange wood gave t h e heaviest shades REPORT
1 Sargent, Chas. S., "Manual of the Trees of S o r t h America." 2 U. S. Dept. of Agr., F a r . Serz'. Czr. 184, "Fustic Wood, I t s Substitutes and Adulterants," by G. B. Sudworth and C. D. hlell. 8 Philadelphia Textile School. New Bedford Textile School. Lowell Textile School, The S o r t h Carolina College of Agriculture and Mechanical Arts. Georgia School of Technology. Osage orange sawdust furnished by Mr. L. C. Bumpus, Farmersville, Teras. 4 Reports on the experiments, giving the results of their test, were submitted b y the coiiperatars, and the information given in this article is in T h e reports are designated the form of extracts taken f r o m these reports by letters, this designation being used to separate the extracts.
