een naar Se EE sates i <= 3 wrivton shuns lapel rea boer re a tay ct ne renee temas =a inayat eee: 5 wh Ni5 ou oY «= — ae = | -- sikh | ee . Digitized by the Internet Archive in 2009 with funding from University of Toronto | ose tee. a ie .e ies ; in 9 as 4 ie i, See “a Tay t y h ee JOURNAL OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS BOARD OF EDITORS C. L. AtsperG, Chairman R. E. DooitrLe C. B. Lipman E. F. Lapp L. L. VAN SLYKE N. A. Parkinson, Associate Editor VOLUME III 1917-20 iy - ual a “a\" | 1920 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS BOX 744, ELEVENTH STREET STATION, WASHINGTON, D. C. S $83 AT v3 Copyricut, 1920 BY JOURNAL OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS CONTENTS PROCEEDINGS OF THE THIRTY-SECOND ANNUAL CONVENTION, NOVEMBER, 1915. MonpAy—AFTERNOON SESSION. PAGE Report on Canned Vegetables. By W. D. Bigelow.....................22005. Reportion Leayand| Coffee) By M> Bartlett... = .:.-- seen amet ds stele 2e et Determination of Saccharin in Foods. By C. B. Gnadinger.................... Report ONUETeSELVAtlVess BYCA Eo CCKED 52 < wens oc oe cee ees | eres. Report on Heavy Metals in Foods. By E. L. P. Treuthardt .................. TuEsDAY—MORNING SESSION. Report on the Separation of Nitrogenous Bodies (Milk and Cheese). By A. W. IBGswoOrthes 52 Geen cae mone ee see eee reece ss cla ohana ajc Glaus sp atti eetAe en sa Report of Secretary-Treasurer. By C.L. Alsberg................--22-+----:- Report on Journal of Association of Official Agricultural Chemists. By C. L. Alisher genera ee ss eee ee een nse ok ce semis = soe vee oe cle cme eels Report of Committee A on Recommendations of Referees. By W.W. Skinner... TuESDAY—AFTERNOON SESSION. Report of Committee B on Recommendations of Referees. By R. E. Stallings... Report of Committee C’on Recommendations of Referees. By H. E. Barnard... Report of Committee on Editing Methods of Analysis. By R. E. Doolittle... ... WEDNESDAY—MORNING SESSION. Repert.on) Phosphoric; Acid« «Byili.05. WiSlKeD......-. vemytvane « oiecer nied eine eiesiceys Report on Use of Citrates in Determination of Phosphoric Acid. By W.J. Jones, jr. Preparation of Organic Material for Determination of Phosphoric Acid and Potash in Aliquots of Same Solution. By R. M. West....................2------ New Method for Drying Ether and Sample in Determination of Ether Extract. BD yERe UIs Wiest he ot ena oe = a os ss eptieseynenecs t posal Sis. Sic Sout ebelayeioets Report of the Committee on Availability of Phosphoric Acid in Basic Slag. By ERRATA. Page 193, Test, second paragraph, line 2.—Omit formula. Page 262, line 9—Change the second formula to read: s ” BiStm 5 G+40+ 000 G? Report of Committee of Review on the Analysis of Lime Sulphur Solutions. By lit ad Olnat Pir iomaonad acknn ach Da BBE rte cb. (Ogee eee rennet ns nao sae Resortion Alkaloids: (By HG atllenseosnctiss coe ema shee sa cess neeenin tees Delicate Test:for Strychnin: “BytHo. Buc... aia... Inorganic Phosphorus in Animal Tissue. By F. M. Beegle...............----- : Report on Nitrogen. By R. N. Brackett and H. D. Haskins.................. Symposium on Determination of Nitrogen in Fertilizers. By P. F. Trowbridge .. ‘ Investigations of the Kjeldahl Method for the Determination of Nitrogen. By PSK Phelps andjEl. Ws Daudt .f08. 5: scccihe J cies ee Resins side reais isl Notes on Use of Potassium Permanganate in Determining Nitrogen by the Kjel- dahl Method. By William Frear, Walter Thomas and H. D. Edmiston... . . Report on Testing Chemical Reagents. By C. O. Ewing.................----: il se one iv CONTENTS PROCEEDINGS OF THE THIRTY-THIRD ANNUAL CONVENTION, NOVEMBER, 1916. PAGE Officers, Referees, Associate Referees and Committees of the Association of Official Agricultural Chemists, for the Year Ending November, 1917............... 228 Members :and' Visitors Present... 32. shee oe at cet ee ee Aol fe 231 President's Address. By R:iNjBracketteesen. cent ee eee eee ne eee cue 238 Monpay—MorwninG SEssIon. Report on Dairy Products.. “By /ifarcryeRlueters..-eee so. ss. o see ses caeee ss 254 Report on Foods and Feeding Stuffs. By A. C. Summers..................... 254 Report on Feed Adulteration. By Carleton Cutler........................... 255 Report on’ Crude. Wiber:) (By iG: Ka Branca ser ooo ent y eee ele ne 256 Report on: Sugar. ByiG@: Ac Browne peepee tee or iacca ee 4 chy te ee ee 261 Recommendations'oniSugar. By, WesDNiome va. +... 0.5040. 70 oes ale Sees 263 Water in Foods and Feeding Stuffs. By W. J. McGee......................-. 264 Inorganic Phosphorus in Animal Tissue. By E. B. Forbes.................... 264 Report on the Separation of Nitrogenous Substances in Milk and Cheese. By L, ‘S53, Palmer 8.568098 ~ «asc ee eye ie ee Cee Oe eg i ae 273 Origin of the Neutralization Precipitate of Cows’ Milk. By L. S. Palmer . . 274 Progress Report on the Separation of Nitrogenous Substances in Meat Products. By P: F:. Frowbridge?:* ;.: |. Pepe eee eee eee eee 275 Report on Testing Chemical Reagents. By C. O. Ewing...................... 276 Monpay—AFTERNOON SESSION. Report’on-PhosphoresAcids | By, Wie GQuesi} Da oe ee ere ee ee 279 Report on Phosphoric Acid in Basic Slag. By C. 8. Lykes...................- 286 Reportion: Nitrogen) By EL. D:. Elaskanis mann tise on ene eee ene ene ee 289 Report on Nitrogen Determination. By R. B. Deemer....................... 299 Substitution of Sodium Sulphate for Potaksiuen Sulphate in the Kjeldahl up Arnold Method for the Determination of Ammonia in Fertilizers. By T. Jared ee Ie oR eA ea eee ne " 304 Investigations of the Kjeldahl Method for Determining Nitrogen. By I. K. Phelps Br Report on Potash: By, T’.19:, Jarrell eee cee en es ene Sera The Separation and Gravimetric Estimation of Potassium. By S. B. Kuzirian.. Sot A Study of the Availability of Potash in Commercial Wood Mena. By R. E. Rose 323 Nitrogenous Compounds in Soils. ByjG a Bolapmanteperee mys seco) occ tee 326 Report on Inorganic Plant Constituents. By A. J. Patten..................... 329 Reportion Insecticides wbyp hts ©. (Rh orc keene een 331 The Occurrence and Determination of Astt and Asv in the Presence of Each Other in Arsenical Insecticides. By Ru Gi Roark. ..2.4...-.5..0 1. kee se 358 Report on Waters SiBy Wen Sinn create nen mens er nee ee en 368 Report on the Lime Requirement of Soils. By F. P. Veitch.................... 371 Report on Synthetic Products. By W.@O; Hmery...........:.........0..0:08 374 Report on Alkaloids: sBy, HC. Hullerseemamen tneemeuitins = tn) unl Annee 379 Report on Medicinal Plants. By Arno Viehoever.................+.-+ seers 381 Report'on"Papainy ‘By VAKo'Ghesnutee eee heen ee kL eee 387 TurspAy—MonrninG SEssron. Report on Food Adulteration. By Julius Hortvet.............00..00e0-ee 00s 398 Report:on Colors. By WE. Mathewson). ...2.... 10, cee eee eee ee 400 Tartrazine.’ By A.M. Doyle Saad hws Ct OLE eee | tee ak 402 Report on Fruit Produc on By P. B. Dunbar and H. A. Lepper................ 402 Report on Wine. By B: G. Hartmann.eo |). seen eee ae eeieaiae aie 409 Titration of Acidity in ( soloed Solutions. By Bi GeHartmann.. if: e202. he ee 410 The Isolation and Identification of G lycerol in Cider Vinegar. By R. W. Balcom and EG: Grabs: os.i eke el ae ee, Le ie Oa te eee 411 A Note on the Calculation of the Volume of a Liquid from Weight and Specific Gravity:' By Ry W. Balcom:.:) 1 Ne ee. ald ee 413 Report on F lavoring Extracts. ‘By Al Raul! ee ee! og... doe ee 415 Address. By G.'S. Vroomani. (ia. oi oe Rte PR aE Le ee IE ne 418 Honorary President's Address. Sy ‘H.W. W | CV RS a LD tN REE SE i SS tec 420 a CONTENTS v TuEspAY—AFTERNOON SESSION. PAGE IRepoumonnspicess oO Yulia Boma: i= 2. 4.5 biyacscee ee ee eens ae mate are aiei eres 427 Repacmony Baking, Powder. By H. iE. Pattem. .\..)..65lecceiene ces citinciea sae os 429 Repontron Hats andiOius ee le 436 ivepore on CerealiProducts) oby.0./A- We! Glercs...... 5 aera s senses ce se 446 Report on Canned Vegetables. By W. D. Bigelow...................--0000-- 453 Report on Cocoa and Cocoa Products. By Eugene Bloomberg................. 486 Report on Lea aud: Goftees by Ela iV; Loomis: 5... 2... 1. = 2sieeiee eccrine cia es ee e-et- 498 Report, on) Preservatives: ab yeAcp Es SEK ED... cs/.ae1s/ see enon cone aa ea de 504 eporton) VWetalsiin) hoods wBy David Klein « <%)i7..s12.0.4, ho eaten tree ae oe 512 WEDNESDAY— MORNING SESSION. Report of Committee A on Recommendations of Referees and Revision of Methods. Loy Wis Ws SST ae Bt ee os ere ein teode 519 Report of Committee B on Recommendations of Referees and Revision of Methods. yale Bir ballin ps te kat et eee os aoe odes Sle ela ise lore Soon shape MRS oe 524 Report of Committee C on Recommendations of Referees and Revision of Methods. LBP Lab Be Lente bee eee ats ats aie ait Gls eee tse 3 rR etch oC A be ae 528 Report of Committee on Editing Methods of Analysis. By R. E. Doolittle...... 537 WEDNESDAY—AFTERNOON SESSION. Report of Secretary-Treasurer. By C. L. Alsberg...................-.--0-00- 576 freporo.onumlhe Journals. iyi Ee vAISberg.. 0. «jie ecisia,s.02 ere raepetoeiererons 578 Report of Committee to Cooperate with Other Committees on Food Definitions andistandards) uby William brea o..\~ 5.25 ALI PUST Io Lop eeectntlec tacit cies -/- 5.62 September23.... S2.ce0s06-i6 4.41 September 4ep sca atecwictacs) = 4.90 INFLUENCE OF RAINFALL ON THE COMPOSITION OF TOMATOES. It is generally believed by packers that rainfall increases the water content of tomatoes to a very material extent. It appears that tomatoes gathered immediately after a heavy rain are more sloppy on the peelers’ tables, and that after canning the amount of solid meat, as determined by a screen, is less than with tomatoes gathered after a period of normal rainfall. It was desired to determine whether this condition was due to a difference in the water content of the tomatoes or to a change in the tomatoes causing them to permit the separation of the juice more readily. This subject has been discussed in part under the preceding topic. It is unfortunate that the rainfall during the seasons of 1914 and 1915 was not so distributed as to permit conclusive data on this subject. In Table 3 are given the results obtained by the examination of 7 varieties of toma- toes grown at the Arlington Experimental Farm and 3 varieties grown at the Maryland Experiment Station during the season of 1914. Making allowance for the variation in individua] samples referred to above, there is a marked tendency in most, if not all, of the varieties mentioned, for the content of solids and acid to be higher during the early part of the season than during the latter part. It is interesting to note in Table 6 that the rainfall was very much higher during the early part of the season than during the latter part. Unfortunately, this work was not begun until about September 1, 1914. The rainfall during the last half of the preceding month was 4.66 inches on the Arlington Farm, 2.66 inches of which fell on August 29. The ground was practically saturated, therefore, when our first samples were secured, and very little rain fell after that date until the end of the season. It will be noted that only 0.70 inch fell on the Arlington Experimental Farm and 0.76 inch fell at the Maryland Experiment Station during September. During the first part of the month of October, the rainfall at the Mary- land Station was no higher—up to the 12th when the last sample from that Station was secured, only 0.34 inch of rain fell. As stated under the preceding topic, we are unable to draw definite conclusions from these results regarding the influence of rainfall on the composition of the to- matoes. The general tendency to lower results as the season progresses may be due to other causes. It was hoped that during the season a heavy drenching rain might come, preceded and followed by dry or nor- mal weather. In this we were disappoinied. } [Vol. III, No. 1 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS 8 er EE “WI8 gy] [ByUeUIedxg uoyBurpry “MIB y [ByUSUTIedxy UOySUTLIy “WIRY [BJUNUTIIOdxY UOYDMIpIY “MIB [ByuUeMIIedxy uogsuT[Ty UNG] [BJ USMIIEdxY UOYSUTLIY UNG] [RJUOUTTIEdxG UOYSUILTY U01784g JusUIIIOdx| puBlAIB] “M017B4g yueuTIIOdx puBpAIBy m0I}B1g JuouTIOdxyY puBlAIByy “101984g JUSMIIIOdx| puBlAIB ‘MO1Ze4g JuoUMIIEdx| puvlAreyy “M01484g JusUITIEdxy puBlAIBy] “W01784g JuotIAOdx| puwlAIBpy 017849 JuoUIIIedx| puvpAIBy “U01784g JuoUTIIedxG puvpAIeyy “ULIBy [BJUeUTIIedx| u0yduT[Iy ‘ULIBY [BJUeUTTIIedxy UOYsUILY “UB [ByUeUMIEdxY WOZdUT]IV “ULIB | [ByUeMIIedxY UOysuTLy ULIGY [BJUeMTIEdxy WOYSUTPIY “UB [BJUOUITIOdxY UOZUTLIW “UB [BjueuIedxy uoysurp1y “ULIB] [B]UIUTIIEdXT UOysuIpLY ‘UB [eyueuiedx | uoysulIy “WB [ByUeUTTIedxG UOIsUTLIy “WB [eyUeUIIedxG uoysuTLIy NMOUD GUTH OF Ly ah 1d 0 OD 1d OD I OD wmode Higa digs Ao at oO HCO OILV4H aiov -uvoos 9°01 col ceil col FMA HMAMOD Nxt 1D ONHH DAHH ADACANDHMR ON ool mone on ihan lanl el mn 1 aane saltos TOP oOo mance ONaaN ONOr oS O19 = Su pT} quad sad €8°1e CFO 8L'T €2°F ey eee <2 ean edi gnq |[BUls I9UIUIN Fy 8 UOJSSUIATT og “qdag LSZ1 coe | OF 0 | 281 | T6°e | °°" edta ‘xourumy suoysBurary | Of “dag | OFCT Tre | 6F°0 | 90°% | Io'F |. odta ‘sour 8, oyssurary | CT “3dag | ICT oz'ze | 99°0 | FI's | S€°h jodia Ajavou ‘roma s,uoysdurary | OT “3deg | FICT eg'ze | ¥9°0 | 91's | 88°F | ** eda ‘rouumy s,uoysBurary | TT “deg | 6801 stool 080 | 29°% | 80°F | edt ‘aourumyAZ s,woysdurary | 1g “Bny | E901 C228, | 8F:0 | LO | ree oper sees | et 900") OBST BORCHa|SOO NO" | pecacs NODE P | a" eon Se oe Se tea ‘par Sofa | § 490 | LGT COrTGs| LOMO MDL MLN erCer | oe ae jyeuus ‘pas Aaqoy | T “390 | 09ZT 7°88 | 6F'0 | 02's | EFF | Pex Aoqteyr | 82 “deg | seer LEEPEUSE IO, IKOCNG ROME GI 5 gee ee Se ae par Aayjey | ¢@ ‘ydeg | ce2r 1L:2e | 27°0 | 812 | 997% |"170 71111 edt ‘pos Aojqoyt | 12 “adeg | Fecr ise | ch'0 | F's | SOF | edt ‘pax Aoyfoyy | 12 “3dag | eect GGreeh| PAO |#20e: | LAiraleamie a Se ao adis ‘par Aaqay | LT “349g | 82a! 0°28) | eotO |MzoleGyer |i we eee oe Ges par Aaqjay | FI “340g | STZT TES TEN GO 7On WOO AT BSB aGalh seo ee Geomen § radia ‘Aqnvag | ct “3dag | 0221 OL Te | 69°0 | £eT | oth | edt “Aaneag | oF “3deg | STCr GZ LE | ¥9'0 | 08'S | Ze S4)" ss adis ‘Aynveg | 7 “qdeg | FS0T shee 9F 0 19'Z 89'S SDS ReiTTOD HOO SEO ATS fa ‘Aynveg Te “any S901 8b 18 | 6f°0 | 8L'T | 60°F eagle ae at het adi ‘yauroD | ct “ydag | Eza1 #18 | 09°0 | G2°T | OFF | 710112 edte fgemm09 | OT “3498 | 80cT Leese | OLON use OGG ss eee ee ee adr “yaur0p | 7 “3dag | 8801 pte 1¢'0 | 00g | 9° | edt Gom0g | Te “Sny | 2901 ose | €4°0 | 16'S | G4 adi ‘amor Ape sqeyO | F “deg | esol e908 | €9°0 | 891 | cee) adi ‘omer Ape syTeGO | & “3des | oFGT a co 6¢'0 | 80'¢ | og | edta ‘Tomer Ajzeg SqIeYO | Te “Bn | FOOT qua2 Jad|juaa sad|juao sad ee 41aje o1}10 Rae 1 8Plos ay fae es Teeny @1dNVS8 dO NOMWdIuosad aivd Son Jeu] goinr aguagLid do NOILISOdHKOO “wospas ay} jnoybnosy) amy 07 ay wosf saojDUWO} fo aoml passaidza fo uoyisodwoy ‘g¢ GIavVL VEGETABLES CANNED BIGELOW 1917] MIB [eJuoUTINdX Ty UO\sUILLy ‘UIB yy [ByUeUIIIedxy UOySuTpIy “WIE YY [BJUSMITIOdXy UOysUI[LV WIG] [BJUsUITIOdxy UOYSUTTLY “WB [BJUSMMIIEdx Ty UOYsUITIY “U014BYg JUeTITIOdxGY puRlAiv yy U019B49 JUOTMIIOdXY puvpAreyy “U01984Q9 JUOUTIOdxy puvlAivyy “U019849 JUOUTTIOdX Ty puBLAIB] “U014B49 JuOUITIOdxX Ty puBlAIB\, "W014 BIG JuoMITIEdx Ty puvpAie py “U014ZBA9 JuOUTTIOdxy puvlAIByy "U014849 JuoITIOdx Ty puBlAIsyy W014B49 JUOTUTIEdxGY puBlAIeyy UIBY [BJUeMIIEdxy UOysUILIY WIG] [ByUSMMTIedxTy UOysUTLALW “ULIBY [BQUOUITIOdxy woOysUTLIW WIE [BJUsUITIEdxXG UOYSUTpIy ‘UNBY] [BJUsMMTIOdXY UOYSUTpIY WIB YB UeMIedxg uoysurpry “UIB,] [BjyUeUIIedxy uoysulpry “UIB Y [eyusuTIedxy UoysUTL IW “WIRY [BjUeUITIedxyY UO sUILIy “WIRY [BjyUeUTIedxy U0ySuTIy “WIG [Bj{UeMIIedxyY uoysuTLy U01YB4G JuoUITIedxY puvlArepy “U014B49 JuoUTTIedxG puBlArepy “U019849 quoUITIedxy puvlArB yy “U01VB49 JusUTTIedxY puBlAIB HOI4B4G JuauITIedxy puvlAie yy “U014849 JuoUTIedxYy puB[AIe yy “WHOA JueUtIedxy puvlAreyy “U014849 JuOUITIOdxy puvlAre yy “UOIVR4g JuoUMIOdxY puvpAis yy OO strip IDHH HHO AID HBOMAHMM OHHANMHOMHHMMMID DAMMANN Lo) piece} on OD eto. NOMNN TOMATHOMNOMYN Onan ~ ~ cr 0 rH TIOr~owmodwnnos SH SO 8D 2D HD 19 19 AD SH HCO 1D 1D DDOMmID DADO SSSSSSSOSS SCSOOSDSDS SOoOSOSCOSoOoOSoCSSCSoSS BREESE 0D 29 19 a NANOMNRANNTAN RAGIN NHN N nN Senn HANNON soo Nw ANN +++ adti Ayreou * adta Ayrvou ee eae adi ‘ONdBA UID, OL 38 ButArp Aq pauruns0jeq 1 make ‘edrr ‘Aydouy, | eT ydag | Zar me es adia ‘Aydoary, | OT 3deg | OIZT ‘sadra ‘Aydory, | 7 “3dag | 2801 “-adia Ayavey ‘Aydory, | g “ydag | S20 ae adia ‘Aqdory, | Tg “ny | 6901 See adia ‘90099 | ZI “490 | 82zT ‘adra Ayarey ‘au0jg | ¢ *400 | 89zT vedia qsnt ‘omoyg | T 490 | I9ZT SEI adria ‘904g | gz “gydag | FezT ag eee adta ‘auoyg | eg “ydag | IecT ie ea adia ‘aug | 1Z “3deg | ZeZI Bora adra ‘au0gg | zt “3dag | OFZT “-adta [NJ ‘au04g | FT “ydag | OTZT “-adra Aparez ‘ou0gg | y “gdag | 2OZT a edi ‘auo4g | gg 3dag | cRzT sesh aaars adia ‘au0yg | GT “3dag | zzzT adir ‘au0yg | # “ydeg | 9801 adi Ayarey ‘au0yg | g “49dag | 920T pee Se adi ‘903g | Te “Sny | egor ‘(pay) Bsorapuog | og 3dag | GezT ‘(pey) Bsorepuog | gz “dag | 6gZI ‘(pay) Bsorapuog | ¢T ‘ydag | gTzt ‘(pey) Bsorspuog | OT 3deg | T1ZT ‘(pey) Bsorepuog | Z “ydag | Tg0T ‘(pey) Bsorspuog | 1g “Sny | 990T ‘ssoTOUBIN | ZI 990 | 6L2T ‘s80[Q98W | $ “390 | 69zT SS9]YOUBIN | T “990 | Z9ZT ‘sso[oyRN | 8c “3deg | 9ccI ‘ssaTqoqyey | oz 3deg | eect ‘SSoTYOUBI | 1G “9deg | LeZT ‘ssoTqoyeyy | 21 3dag | Seer ‘ssoTqoqeyy | FI “3deg | ZIZT ‘ssaTqoyey | 2 “349g | 90eT 10 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 In 1915 the work was repeated, except that only solids and index of refraction were determined in the juice of the tomatoes and only one variety was employed. This time the work was begun with the first TABLE 4. Composition of expressed juice of are ve time to time throughout the season ] STONE VARIETY, GROWN AT ARLINGTON STONE VARIETY, GROWN AT MARYLAND EXPERIMENTAL FARM! EXPERIMENT STATION > pierre = ; # eee Barolo’ |” ate > | Berean | ereeee,1||) Beers) 108 Bowel 6 ees 17.5°C. 17.5°C. 2060 Aug. 6 5.32 36.05 2099 Sept. 7 Deak 36.19 2061 Aug. 10 5.66 37.25 2313 Sept. 9 5.35 36.19 2062 Aug. 13 5.31 35.70 2316 Sept. 11 4.83 33.65 2063 Aug. 17 5.14 35.44 2319 Sept. 14 5.20 35.50 2065 Aug. 18 5.49 36.76 2321 Sept. 16 5.10 35.00 2066 Aug. 20 5.13 35.51 2322 Sept. 18 4.49 32.50 2090 Aug. 23 5.04 34.86 2324 Sept. 21 4.56 33.49 2091 Aug. 25 4.62 33.49 2340 Sept. 23 4.62 33.08 2093 Aug. 27 tay le/ 35.15 2346 Sept. 28 5.09 35.57 2094 Aug. 30 5.15 35.47 2348 Sept. 30 4.87 34.22 2096 Sept. 1 4.84 34.65 2352 Oct. 2 4.95 34.42 2098 Sept. 3 5.19 36.02 2366 OctaeD 4.75 33.60 2312 Sept. 8 4.96 34.52 2369 Oct. 7 5.19 35.62 2315 Sept. 10 5.14 35.85 2371 Oct. 9 4.76 33.65 2317 Sept. 13 4.19 31.15 2373 Oct. 12 4.61 33.50 2318 Sept. 13 4.57 32.45 2374 Oct. 14 5.08 33.80 2320 Sept. 15 4.32 32.20 2395 Oct. 16 4.51 32.37 2323 Sept. 20 3.94 31.05 2396 Oct. 19 4.41 32.50 2344 Sept. 27 4.28 32.22 2400 Oct. 21 4.85 34.22 2 2403 Oct. 23 4.97 35.00 * 2404 Oct. 26 4.23 31.80 1 Samples were taken from ripe tomatoes. 2 These samples being the last of the season were not entirely normal. They were badly shaped and not quite ripe. SAMPLE No. All other samples were normal and ripe. DATE TABLE 5. : Composition of expressed juice of tomatoes throughout the season, grown on a single vine (1915). {Stone variety, grown at Arlington Experimental Farm.] DESCRIPTION OF SAMPLE Aug. Aug. Aug. Sept. Sept. Sept. Sept. Ripe Nearly ripe, sun-burned......... 1 large, 3 small fruit ripening from the vines set aside for the work. season was very unfavorable to tomatoes. PER CENT IMMERSION RE- cone | PRACTONEEE® 4.97 34.53 4.77 33.60 5.13 35.07 4.73 33.94 4.54 32.97 4.60 33.27 4.88 34.82 Unfortunately, the Because of the heavy rain- fall and cold weather the fruit did not begin to ripen until two or three weeks after the customary time. The results are given in Tables 4 and 1917] BIGELOW: CANNED VEGETABLES itil 5. Here again the climatic conditions are such that we are not warranted in drawing definite conclusions regarding the relation between the composition of the tomatoes and the amount of rainfall. By refer- ring to Table 6, it will be noted that the rainfall was much heavier in the month of August than later in the season. At the Arlington Experi- mental Farm the tomatoes ripened first on August 6, and they continued to ripen on the same vines until September 27. In the first six days of August, Just previous to the date of the first sample, 3 inches of rain fell at Arlington and the ground was thoroughly soaked. TABLE 6. . Rainfall (in inches) at Arlington Experimental Farm and Maryland Experiment Station, 1914 and 1915. 7 1914 1915 ARLINGTON EXPERIMENTAL] MARYLAND EXPERIMENT) ARLINGTON EXPERIMENTAL|MARYLAND EXPERIMENT FARM (AUG. 15-sEPT.30) | sTATION (SEPT. l-ocr.15)|| FARM (auc. 1-sEPT.30) |sTaTION (AUG. 15-ocrT. 31) Day of month | Inches | Day of month] Inches | Day of month} Inches | Day of month| Inches Aug. 22 0.06 Sept. 3 tr. || Aug. 2 | 0.05 Aug. 17 tr. Aug. 25 | 0.40 | Sept. 8 tr. || Aug. 3 | 0.02 Aug. 20 tr. Aug. 26 1.00 Sept. 12 0.35 | Aug. 4 2.52 Aug. 21 0.35 Aug. 27 0.01 Sept. 25 | 0.41 | Aug. 5 ! 0.35 Aug. 22 | 0.05 Aug. 28 0.40 Oct. 5 0.16 Aug. 6 0.06 Aug. 25 0.05 Aug. 29 2.66 Oct. 8 0.18 Aug. 7 0.87 Aug. 28 0.74 Aug. 30 0.13 Oct. 15 0.24 Aug. 10 0.10 Sept. 2 tr. Sept. 12 0.26 | Aug. 11 0.14 Sept. 5 tr. Sept. 25 0.44 | Aug. 12 1.47 Sept. 6 0.60 | Aug. 13 0.03 Sept. 7 0.01 | Aug. 16 | 0.06 Sept. 12 tr. | Aug. 21 0.48 Sept. 19 tet Aug. 23 0.08 Oct. 4 tr. | Aug. 25 | 0.01 Oct: 55 0.43 | Aug. 28 | 0.35 Oct. 6 0.13 Aug. 30 1.62 Oct. 7 0.21 Sept. 6 0.63 Oct. 8 0.23 | Sept. 7 | 0.12 Oct. 14 | 0.04 Sept. 18 0.12 Oct. 16 0.22 Sept. 19 0.49 Oct. 20 0.42 Sept. 21 0.36 Oct. 27 || 0.22 The amount of rainfall at Arlington in successive periods during the summer and the solid content of the tomato juice during the same periods are shown in the following tabular statement: Relation of rainfall at Arlington Farm to solid content of tomato juice (1918). AVERAGE DATE RAINFALL SOLIDS IN TOMATO JUICE inches per cent REID UB pL LOY wonton ici c Saks os oat etatia ae onSoe ca eto 5.61 5.43 ENTE TEMES ee See SOC e aE Eare are 2.60 §.11 DEBLEM DELP —l oer ea cae aetalae ohyarare,s a chsie sate 0.75 4.74 DEpLembers1G-GU wees aes Nee ae ncoe we Se ee 0.97 4.11 Pripiintgl 3 Beet pice Ae ee NRE Rh oo Bk 8.21 5.20 DER UGIIELeL— SU eee a en anh res cisiabe Sea ea Tae 4.60 \ 12 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Very similar results were obtained from tomatoes grown at the Mary- land Experiment Station. At this place they ripened later, the first fruit being ready to pick on September 7. During the first 7 days of this month the rainfall at the Maryland Station was 0.61 inch, and 0.74 inch fell on August 28. The ground was not as wet as at the Arlington Farm when the first tomatoes were picked at the latter place. Owing to the fact that the fruit ripened later, there was not so great a difference between the rainfall at the beginning of the season and later in the season at the Maryland Experiment Station as at the Arlington Farm. Notwithstanding this, there was a marked diminution of solids as the season progressed, though not so great as at the Arlington Farm. This suggests the probability that the diminution of solids was due, at least in part, to some other cause than variation of rainfall. The amount of rain- fall and the content of solids in the juice of the tomatoes at the Maryland Station is shown in the following tabular statement: Relation of rainfall at Maryland Station to solid content of tomato juice (1915). ‘ AVERAGE DATE RAINFALL SOLIDS IN TOMATO JUICE inches per cent September 1=15 20% occ 0b ae nin scl eerone eeee 0.61 5.13 September 16-304. .f:s ice ate semen een ie ae 1.11 4.79 October leo etiscie.tecelee os eee eee etentee ney: 1.04 4.89 October 16=272)25.5...20 SR eee oat eee te ceieneiee 0.86 4.59 September! 1-30.48 cfs etavst ator eee eloelertele 1.72 4.94 Qetober 1-27.57 2265. edteloh: cine tates eo tocar: 1.90 4.75 In Table 5 is given the solids content of a number of samples of tomatoes taken at successive times of ripening from a single vine grown at the Arlington Experimental Farm. These results show the same individual variation which has been pointed out in the preceding tables. Taking the figures in this table as a whole, there appears to be a tendency to a lower percentage of solids as the season progresses, although this tendency is not as marked as is shown in Table 4. It is pointed out above that the solids content is higher in the tomatoes ripening early in the season than those ripening later in the season. As already stated, the rainfall during the last two years was also much heavier shortly before they were picked than later in the season. At the same time, as stated above, the thought suggests itself that this higher content of solids may be due to the greater strength of the vines at the beginning of the season than later. Moreover, the tomatoes ripen better during the hot weather that prevails in August than during the cooler weather in the fall. On the whole, the results do not warrant definite conclusions, and it is hoped that during the coming summer results may be obtained for a period of time which includes a drenching rain preceded and followed by dry weather. 1917] BIGELOW: CANNED VEGETABLES 13 COMPOSITION OF TOMATOES FROM BLIGHTED VINES. The tomato crop is sometimes attacked by leaf blight (Septoria lycoper- sici). The blight usually makes its appearance before the fruit has rip- ened, and if climatic conditions are favorable it progresses rapidly after the ripening of the fruit begins. If rainfall occurs at about that time, it is followed by an especially rapid growth of the blight. The result is the destruction of the leaves of the plant and the imperfect development of the fruit. Fruit that is practically grown and nearly mature before the vine is destroyed by blight takes on the normal red color, and practically the flavor of ripe tomatoes. Fruit that is only partly grown takes on an external color of yellow or red, according to its state of maturity. When such fruit is cut the appearance of the section is widely different from that of ripe fruit. The color is pale pink or salmon, and the outer wall and the dividing walls between the seed receptacles are thin and pale. The composition of this blighted fruit is practically the same as that of green fruit at the same stage of maturity. Such fruit is usually rejected when offered for sale, but the dividing line between mature and immature ‘fruit is hard to fix. The composition of the fruit on blighted vines is shown in Table 7. Each sample in this table consisted of at least six fruits. Notwithstanding this, the variations of individual samples are great, and the table is only of value because of the general tendency of the results it gives. The first three samples are of special interest. No. 1091 was nearly ripe when its development was stopped by blight. It, there- fore, closely resembled normal ripe fruit in color, thickness cf walls be- tween seed receptacles, and composition of juice. Samples 1090 and 1092 were checked by blight while less mature. Their color was pale reddish yellow, the walls between the seed receptacles were thin, and the compo- sition of the juice resembles that of green tomatoes. Samples 1093 and 1094 were different portions of the same tomatoes taken from blighted vines. They were nearly mature before their growth was checked by blight. The lower halves of the fruits were the color of normal ripe to- matoes. The upper (stem) halves were badly sunburned and probably did not color for that reason. The remaining samples in Table 7 were taken near the end of the season—over a month after the fruit had begun to ripen in the patch. The plants were badly attacked by blight when the fruit began to ripen, but were not destroyed. ‘The plants retained some leaves until after October 7. They bore fruit throughout the season that was normal] in color and almost normal in form and appearance, though the yield was light. The fruit was separated according to size in preparing samples for analysis, as is indicated in the table. From the color of the samples and the thickness and color of the dividing walls between the sections of [Vol. III, No. 1 14 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS a ———————_——_—___———_———_———_—————_———— "PY ‘20BID Op WAG “PIN ‘8981 Op OAR “PIN ‘90BID OP OIAB FT “PY ‘20815 op ABT “PIN ‘998. Op O1AB AL Aesi0f MIN, Aosier MON Aasior MON Aasief MIN Aasior MON, g@ounos 8G (aes O€ LG o> OlLya uvoeos ~diov £9°€ 61h 16°€ 60°F eZ oo 7 Ie'P We £0°€ 68° € quad sad |juao sad GSI | GLE | 68°6c | 8F'0 | SET T GI | 6 8€ | 62° TE | 19°0 | €9 T GGL | OLE | OZ TE | 6F'0 | ZHI 9 €1 | O98 | 6S’ TE | SSO | ZF T 6° CE | 2 88 | 22° 1S | §9°0 | Zo T TAD | CELA || £55200) | Sel Gee QT 2 OFa)| eam ee | OhE OM OOnG ¥ ST | SSF | ST 62 | 4h'0 | SST SFI | 9 EF | 08 92 | SFO | GET IPI | 2°GS | SO'TE | Ih'0 | 90% quao vad|yuao wad quao sad ONOBA UI Splfog GoInt GAUALIA dO NOILISOdNOD esveie ele oreo erase eae OT ODN BELLO ‘JoJOUIVIP UI SoyOU! Z UeYY sso] ‘SseroyByY | 2 “490 GLZ1 eter tee eee eect eteeeseessss39109 [BUL ~10U ‘19POUIeIP UI soyoUT g 04 Z ‘ssorqoqeyy | 2 990 PLZT Si pansogpannds "39100 [RUIOU ‘JoJOUIIP UT seqour Z UBYY Sse ‘sseTyoyByy | 2 400 | 10SET RAROONNORE pid ado pu Eoodadt “10[09 [wut ~10u ‘JeyoureIp Ul SOYOUT § 0F Z ‘ssoTqoqyvyy, | 2 990 | 16FST yobcoduNoHS Gatos mnaas ee Osan “1009 [RUN tou ‘JaJOUIBIP UI sayouUT g BAO ‘SSaTyOIVY | L 39O | iSFST E601 “ON Jo jeg soddn ‘10]Oo usei3 ‘peurnquns ‘ez18 wInTpour seq ermuog | g¢ 4dog F601 srs ones es 18809810} JO F{BY 1eMO] “IO]Oo jeuiou ‘azIs uINTpeut ‘980 aruuog | ¢ 4dog e601 : ToePioddoOuay coat) oMoU GELS “prop Aypeo1yovid saavoy] ‘s[[eA arya ‘10]09 pel ~YSTMoT]od | ‘UMOId piIq4- -ou0 ‘qseq elmu0g gydag Z601 J aseans did oF aq puDceBSe ie “epRaDp AypBotqovid Soave] ‘s]]eM arg} ‘10]00 ped ~ystmoryos ‘aMols J[By-9u0 “ysaq orauog ¢ ‘ydag 060I verses sngap ATBorjovid SOAGOT ‘IO[00 [euliou ‘ezIs WNIpeu MOTeq “yseg eruuog | ¢ “ydag 1601 @1aWY8 JO Nolddosaa aLva Pas ‘saura pajybug worl seojnwo} fo uoyrsodwog *) O1av 1917] BIGELOW: CANNED VEGETABLES 15 the seed cavities, as well as the analytical results given in the table, it appears that the small tomatoes were scarcely, if at all, less mature than the large. COMPOSITION OF CANNED TOMATOES. As stated above, the juice expressed from the flesh of tomatoes is widely different in composition from that of the seed receptacles. The percentage of sugar is much greater in the former and the percentage of acid is greater in the latter. In peeling tomatoes for canning, the seed TABLE 8. Composition of juice expressed jrom duplicate cans of tomatoes. [Date: October 7, 1914.] SOLIDS REFRAC- sotips| cALcu- |TION INDEX BY LATED OF SEE DRY- FROM FILTERED Vee hose Inc! REFRAC- JUICE TION INDEX| (17.5°C). per cent) per cent 1546 5.20 35.77 |\Great B. B.—14 to 2 inches in 5.30 36.20 diameter. 1547 an 1-18 Bonnie Best; small. 4.22 31.77 |)Matchless; from blighted vine 1548 4.63 33.46 unsprayed, over 3 inches 4.63 33.46 in diameter. 4.17 31.59 Same as 1548; 2 to 3 inches in 1549 3.90 30.52 dinminter 4.20 | 31.72 emer (RT) INT ae eee 370 | soe | 30.68 Serie Be 1 Minder eee (ots Mle 2 GH. Biri CSkS7) lao. 40) (|) UD clameter: 7 abn Toe aeee 5.84 5.84 38.45 1551 | LD nae ter aiaereciond 5.37 5.56 37.39 |}Large mature tomatoes. Cuetiee cach os 5.37 5.49 37.03 1 Determined by drying at 70°C. in vacuo. receptacles are torn open more or less and the juice runs from them freely. It is probable that the juice that separates on the peeling table comes from the seed receptacles to a greater extent than from the flesh. This is especially probable if the tomatoes are not fully mature. Partly for this reason the composition of individual cans varies even in fancy hand- packed tomatoes, where the cans are filled entirely with the solid meat of the tomato without the addition of any of the juice separated on the peeling table. Moreover, the tomatoes are filled into the can as indi- viduals, and only a small number of tomatoes are placed in a single can. 16 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Even if the tomatoes could be placed in the can without peeling and con- sequent loss of juice, the contents of the individual cans would naturally vary in composition as much as the individual samples in Table 3. Where the cans are simply filled with the tomatoes without pressure, and the interstices are filled with the juice separated in the peelers’ pans or pails, it is apparent that the composition of individual cans may vary consid- erably. This is illustrated by the data given in Table 8, in which is shown the composition of a number of duplicate cans from the same lot of to- matoes. Even these cans do not represent the average commercial pack. They represent a pack put up in the presence of representatives of the laboratory. These samples were much more uniform than the individual cans of an ordinary commercial pack. The tomatoes were pressed into the cans by hand until sufficient juice separated to fill the interstices. None of the juice separating on the peelers’ tables was added to the cans. It will be noted that even under these circumstances there is sometimes a variation between duplicate cans of as much as 10% of the amount of total solids present. STUDY OF THE DRAINED SOLIDS IN CANNED TOMATOES. This subject was studied by the previous referee, who recommended that the investigation be continued. From time to time for a number of years canners have attempted to define the grade of canned tomatoes ordinarily known as standard. Such attempts have been uniformly unsuccessful. A number of years ago three State canners’ associations adopteda standard depending on the weight of tomatoes remaining on a sieve witha quarter-inch mesh. Un- fortunately, this standard was not based on experimental work. When the attempt was made to put it into practice, it was found to be without value, and it has never been found practicable in the trade. The North Dakota Experiment Station for some years has used a modification of this method, the following details of which were given the referee by Mr. Ladd: We first pour the tomatoes upon a sieve, }-inch mesh, as prescribed by the method usually employed, and then the juice that passes through this, with the fine par- ticles of tomato, are passed through a cheesecloth. A small square wooden frame, with sharp-pointed nails standing up at each corner, stands on the top of a good- sized beaker, the cheesecloth sagging somewhat in the center. The juice is poured on this and allowed to run as long as it flows freely, then the cheesecloth is lifted off from one end of the frame so as to pass the solution off from the clogging por- tions; it is then lifted from the other side of the frame back and forth, and allowed to drain as long as it freely drips without applying any pressure whatever, only the slight pressure that comes from the raising of the cheesecloth on opposite sides from time to time to move the juice forward. 1917] BIGELOW: CANNED VEGETABLES 17 The time can hardly be fixed, as tomatoes will differ; those that are overripe, or overcooked, or which have been shipped for a long distance, will not filter as rapidly as those less ripe or not cooked as much, or which have not been subjected to long- distance shipping. It is claimed for this method that the cheesecloth retains the small particles of tomatoes which are separated in storage, shipment, or over- cooking, and thus corrects, in large measure, the inaccuracy of the ordi- nary method. TABLE 9. Drained solids in canned tomatoes (laboratory pack) using screen with one-fourth ; inch mesh. STORED 38 FROZEN 7 SHIPPED DAYS AT canes. ere ear phere per ATURE per cent per cent per cent per cent per cent 8minutes at 212°F. in rotating 719.7 69.4 75.2 BED see eregrereis:6 cooker. 80.1 61.8 74.0 (Ae eaciennaor 78.1 63.7 77.7 Dei Nevate ctorate 30 minutes at 212°F., not cooled. . 68.0 41.7 Sle 79.1 56.5 60.9 60.6 , 47.1 70.5 61.2 59.9 53.4 41.9 (AL Hee Socnoe 30 minutes at 212°F., cooled....... 73.4 59.9 62.2 TB28) ||| Rksccesjccers 74.5 59.8 60.5 ili Bw \Weeeeee arte 73.2 54.3 (PIP? (455) ieeeetces 45 minutes at 212°F., not cooled... 70.6 54.2 54.7 72.1 63.5 65.0 54.9 46.0 66.9 66.7 64.7 47.9 59.2 Onl 70.4 45 minutes at 212°F., cooled...... GOR IIE ere erepstarers 58.9 UE \Rogoacdads C/K eee ete 59.2 AO S2i ll egee sates eee (Asa (| Saebaoer 59.8 MZEOM Ee coeeiste 20 minutes at 225°F., not cooled. . 74.6 61.8 69.8 73.8 60.7 13.2 64.4 61.0 78.3 66.8 77.0 58.1 65.2 74.8 62.7 20 minutes at 225°F., cooled...... 76.2 52.0 70.2 Gielen Prepares averere U2 60.0 65.2 OOP? |e sxote-s 78.4 59.0 51.1 (Bas leaonoacee To study this point, two sets of samples were secured and subjected to different conditions of temperature and storage, including shipment from Washington to Lawrence, Kans., and return. The samples were prepared in the laboratory, and care exercised to secure uniformity. The results are given in Tables 9 and 10. Unfortunately the tomatoes from which these samples were prepared were so green that the canned samples were scarcely merchantable. The tomatoes were packed solid, no juice being used. After sealing, the cans JOUR. ASSOC. OFFICIAL AGRIC. CHEMISTS, VOL. III, NO. 1 18 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 were divided into several lots which were sterilized by different methods, as shown in the tables in the column headed “Process.” The various lots were then treated as stated below, and the percentage of drained solids determined by means of both a quarter-inch screen and cheesecloth as described above. These tomatoes were then treated as follows: Frozen at 0°F.—The samples were placed in cold storage immediately after being canned and held at the temperature of 0° F. for one week. TaBLeE 10. Drained solids in canned tomatoes (laboratory pack) using both screen and cheesecloth. STORED 38 FROZEN 7 SHIPPED DAYS AT STACK PROCESS UNTREATED| DAYs aT |1,300 MILES| WINTER ScENED oor. BY FREIGHT| TEMPER- ATURE per cent per cent per cent per cent per cent 8 minutes at 212°F. in rotating 84.5 75.4 84.9 82.1 ||. cooker. 83.1 70.9 80.8 2920) llr 81.9 71.2 84.1 CY fe) Wado cc 30 minutes at 212°I'., not cooled. . 15.2 57.7 68.1 85.0 65.2 69.7 67.2 58.7 78.7 68.5 64.3 58.7 53.1 TST) Nbr ase foe 30 minutes at 212°F., cooled...... 78.0 66.1 75.3 82.6) |. Sccneee 79.0 66.0 74.7 82:8: ile themes 77.6 63.8 80.9 80:5) excite 45 minutes at 212°F., not cooled. . 77.0 60.8 67.9 78.1 69.7 70.8 61.1 64.8 76.0 71.5 72.1 55.5 74.5 82.3 75.7 45 minutes at 212°F., cooled...... TORO Deets (2.5 80.1 ie eee Sota ts een 70.3 77.3) ||. Reese Bae a Memes 74.9 80.6. ||.aonemes 20 minutes at 225°F., not cooled. . 79.5 68.6 78.8 84.9 67.4 80.7 69.6 69.4 84.2 72.6 96.4 64.9 Uae 82.2 68.6 20 minutes at 225°F., cooled...... 81.4 61.0 79.3 82:6. |G. aeeeees 82.1 69.1 78.0 78:8: || fognewewe 82.8 63.3 59.8 78 4a Weertucenmte Stored 38 days at winter temperature.—The samples were placed out of doors for 38 days during midwinter. During this period the temperature varied from 18° to 64°. The temperature went below 28° thirteen times and below 26° ten times. Stack burned.—The samples were placed in an oven in the laboratory and held at a temperature of from 60° to 70°C. for ten days. This was done to imitate the practice of some canneries of storing cans in a solid stack immediately after proc- — essing and without giving any opportunity for cooling. On account of the unripe condition of the tomatoes discussed in Tables 10 and 11, it was feared the results might not be representative of a 1917} BIGELOW: CANNED VEGETABLES 19 commercial pack, and the work was repeated with a sample of tomatoes packed commercially from fully ripe tomatoes. This sample was also divided into lots which were heated for various lengths of time in addi- tion to the processing given them at the factory. The details of the treatment given the various lots is shown in Table 11. In the selection of both samples above, all possible care was taken to obtain a uniform pack. It is to be expected that considerable variation will occur in the TABLE 11. Drained solids in fancy hand-picked tomatoes (commercial pack). SOLIDS SOLIDS on 4-INCH TREATMENT IN ADDITION TO PROCESSING AT PACKING HOUSE ON }4-INCH |SCREEN AND SCREEN CHEESE- CLOTH Untreated). 3-5 -s.cscese 5 ss We lari Bio c cles Oa tscie's icra eevee else Os 63.2 67.1 60.7 66.1 66.9 70.7 Heated in boiling water 45 minutes, not cooled................. 70.1 74.8 72.2 76.6 67.5 72.3 Heated in boiling water 45 minutes, cooled.................... 73.9 Wee 75.4 78.8 74.5 78.3 SCG [DTT bs oS SAS OSDS REUSE SOS GRO r Dee een EDCcE Sans a caaenranr 57.1 60.9 61.1 65.1 59.9 64.8 Pippedsel, SOU miles Dy EXPFESS oo) Ss Jet rs giaees 3 scloste eRe 6 70.5 75.0 70.8 75.0 71.4 75.6 TGwE Meas yS abi OcK mca aeitecscis os seis «vs ccisemes sacra semeesaens 62.2 65.7 57.3 61.4 51.9 56.9 Stored 38 days at winter temperature...............-.-........ 70.3 75.2 72.4 78.5 W125 76.8 individual samples examined in this manner, and it is interesting to note that this variation is much less in the determinations made with both quarter-inch screen and cheesecloth than in those made with the quarter- inch screen alone. The results shown in these tables are more uniform than can be secured with miscellaneous samples representing different lots of tomatoes, and even different localities and different portions of the season in the same locality. It would be interesting to study the influence of the variety of to- matoes, the degree of maturity, the amount of rainfall, the locality in 20 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 which the fruit is produced, the mode of scalding, and the sizes and shape of pails and pans in which the peeled tomatoes are placed by the peelers. The figures given in the accompanying tables, of course, apply only to “solid pack” tomatoes put up with the utmost care. The variation in the amount of solid meat is, of course, much greater in tomatoes packed under commercial conditions—especially when filled by machinery. COMPOSITION OF TOMATO PULP AND METHODS FOR ITS ANALYSIS. Tomato pulp is now made on a very large scale for the manufacture of ketchup and soup. It is also being packed in increasing quantities in small containers for household use. If sold as tomato pulp or under any similar name without qualification, the product is supposed to consist of the fleshy portion of the tomato separated from skin, cores, and seeds, by means of a fine-mesh screen and suitably concentrated by evaporation. If manufactured from trimming stock in connection with the canning of tomatoes, that fact should, of course, be stated on the label. During the last year a considerable number of samples of pulp of known origin, including samples manufactured in experimental runs conducted by the laboratory of the associate referee, were carefully studied. As a result of this work, it was found that in pulp manufactured from whole tomatoes a very exact relation exists between the results obtained by the following determinations in the pulp and in the filtrate obtained by throw- ing the pulp on a folded filter: Total solids as determined by drying in vacuo at 70°C. Total solids as determined by drying four hours (after apparent dryness) under atmospheric pressure at the boiling point of water. Specific gravity of the pulp. Specific gravity of the filtered liquor. Index of refraction of the filtered liquor. As a result of the data secured from this work, the following relations have been established: DM = Solids by drying at atmospheric pressure X 1.085. = 8’ X 1.12. S = 228 (d — 1.000) + 19.1 (d — 1.015). = 257.5 (d’ — 1.000). = 0.289 (r — 15) — 0.0185 (r — 26.4). = 748 (Np — 1.3332) — 25.5 (Np — 1.3376). = Solids by drying at atmospheric pressure X 1.125. S’= 230 (d’— 1.000). S’— 0.258 (r — 15) — 0.0165 (r — 26.4). 8’= 666 (Ny — 1.3332) — 20.7 (Np — 1.8376). iS) Ss S s/ 1917] BARTLETT: TEA AND COFFEE 21 In which the symbols have the following significance: S = Per cent solids of pulp determined by drying in vacuo at 70°C. S’= Per cent solids of filtrate determined by drying in vacuo at 70°C. d = Specific gravity of pulp at 20°C. = Specific gravity of filtrate at 20°C. r = Scale reading of filtrate on immersion refractometer at 17.5°C. N, = Index of refraction of filtrate at 17.5°C. on Abbe refractometer. From the specific gravity of the filtrate at 20°C., the per cent of solids of the pulp (not of the filtrate) may be ascertained from the Windisch wine table (U. S. Bur. Chem. Bul. 107, Table V, pp. 218-220). The figure 0.05 should be deducted from the percentage of solids given in that table. The solids in the filtrate may be ascertained from the index of re- fraction, using Wagner’s table for beer and wine extract. This table is applicable without correction to the juice of fresh or canned tomatoes. When applying it to the filtrate from pulp of the usual concentration, the figure 0.17 should be deducted from the percentage of solids given. If the product has been salted, the sodium chlorid should be determined and & corresponding correction made in refractive index. The data given above has been arranged in tabular form, but a de- tailed statement is not included in this report, as it has already been published.! No report was made by the associate referee on cocoa and cocoa products. REPORT ON TEA AND COFFEE. By J. M. Bartuert (Agricultural Experiment Station, Orono, Me.), Associate Referee. Work on these materials has again been confined to methods for the determination of caffein. This subject has been considered by the asso- ciation for four years without much progress being made. This is partly because only a few of the members are particularly interested in the de- termination and but little céoperation could be obtained, and partly because nearly every man working upon these materials has some favor- ite method for determining caffein, thus furnishing quite a number of methods to test out. This year the many different methods have been studied to learn which is a practical and accurate method applicable to both tea and coffee. The caffein of coffee is chemically the same as that of tea and, being entirely soluble in hot water, alcohol, and chloroform, can be extracted from either tea or coffee by any one of these solvents. Allen? uses boil- 1 J. Ind. Eng. Chem., 1915, 7: 602. 2 Allen’s Commercial Organic Analysis, Vol. III, Pt. II, p. 490. 22. ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 ing water for both tea and coffee, and states that he has found alcohol to effect no quicker separation and it removes a larger amount of chlorophyll. Paul and Cownley! dried the moistened powder with magnesia and then extracted with aleohol. Hilger and Fricke? extracted caffein from coffee with boiling water, as did Dvorkovitsch and Stahlschmidt from tea. Sullivan in his work on coffee used boiling water, and Fuller in his method uses acidulated boiling water to extract caffein from both tea and coffee. Gorter uses chloroform and a Soxhlet extractor to extract caffein from coffee. With two exceptions the methods mentioned cause a complete extraction of crude cafiein in both materials with hot water. A large volume of liquid, which must be later removed, results from filtering and washing the residues. In the method proposed this difficulty is obvi- ated by using graduated flasks, passing the solutions through dry filters, and using aliquot portions for subsequent work. Several different methods of separating and purifying the crude caffein after extraction are used, most of which give good results when properly carried out, but some are long and tedious, and in the experience of your referee give no more ac- curate results than some of the shorter and less tedious methods. Stahlschmidt’s* method, slightly modified, has been before the associa- tion for four years, and as now presented has been changed only in minor details. The chief changes are the use of dry basic acetate of lead in- stead of the normal to clear the solution, and hydrogen sulphid to re- move the excess of lead instead of sodium phosphate, which gives a pre- cipitate that is very liable to run through the filter. Graduated flasks and aliquot portions are used instead of attempting to wash precipitates, which is a difficult process and incomplete with these materials. The method as recommended is as follows: Weigh 3.125 grams of the finely powdered material, fine enough to pass through a 40-mesh sieve, into a 500 cc. flask, add 225 cc. of water, attach a reflux con- denser, and boil for 3 hours. Add 2 grams of dry basic acetate of lead (Dr. Horn’s) and boil 10 minutes, cool to room temperature, transfer to a graduated 250 ce. flask, make to the mark, thoroughly mix and filter through a dry filter, measure 200 cc. of the filtrate into a 250 cc. graduated flask and pass H,S through it to remove lead. When lead is all precipitated, make the solution up to the mark and filter through dry filter. Measure 200 cc. of this filtrate, representing 2 grams of the original material, into an evaporating dish and concentrate on a steam bath to 40 ce. Wash the concentrated solution with as little water as pos- sible into a small separatory funnel, and shake out four times with chloroform, using 25, 20, 15, and 10 ee. When extracting coffee, make the solution slightly alkaline with ammonia before extracting with chloroform. 1f any emulsion forms, break it up with a stirring rod and run the separate portions of chloroform 1 Allen’s Commercial Organic Analysis, III, Pt. II, p. 491. ?U. S. Bur. Chem. Bul. 107 (rev.), p. 153. * Allen states that this method is applicable to the determination of caffein in coffee as well as tea (Allen’s Commercial Organic Analysis, Vol. III, Pt. II, p. 491). 1917] BARTLETT: TEA AND COFFEE 23 through a 5 cm. filter paper into a small tared Erlenmeyer flask. Remove the — chloroform, dry the residue to a constant weight at 100°C. Divide weight by x35 for percentage. If the caffein is not in pure white crystals determine the nitrogen by Kjeldahl or Gunning methods, and multiply the amount of nitrogen found by factor 3.464 for caffein. Usually the caffein is sufficiently pure, as determined by weight, and seldom varies more than one-tenth or two-tenths per cent from the N determination. The caffein can be estimated after precipitation by iodin solution, as given in Fuller’s method published in the Journal of the Association of Official Agricultural Chemists, volume 1, No. 2, page 203, omitting, however,! the use of animal charcoal, as that retains caffein. Or the coloring matter can be removed by dissolving the caffein in hot water, making alkaline with ammonia, and extracting with chloroform, This requires less time than purifying with iodin solution. For the coéperative work this year a quantity of black tea and a good grade of coffee, Mocha and Java flavor, from the material used in 1914, was ground fine enough to pass a 40-mesh sieve, thoroughly mixed, and “stored in tight cans. The results obtained from two laboratories besides my own are reported in the following tables: Caffein determinations on tea and coffee. TEA. FULLER STAHLSCHMIDT STAHLSCHMIDT METHOD METHOD MODIFIED METHOD N X 3.464 N X 3.464 per cent per cent per cent E per cent per cent H. H. Hanson, Maine Experiment Station, Orono, Me............. 2.70 3.09 2.74 2.91 2.84 13.24 2.91 2.97 2.94 H. C. Fuller, Industrial Institute, Washinton On @ocsssee ce eon: 2.57 Des OLE TN lrxccetcparsrnt | oretelatacl=stal ean spears H. J. Wichmann, reported by P. B. Dunbar, Bureau of Chemis- Us, WWESLIT Grainy 1D) CLES Geis ce| [Soa eels] PSs eters! (ieee 3.01 2.63 3.05 2.53 J. O. Clarke, reported by R. E. Shimiitioss. Wi bins (Crs. 2 ena lesa seoces Heaece er ees meee 2.72 2.60 : 2.63 | 2.50 J. M. Bartlett, Maine Experiment MoatlONeOrOnOy Mert ern eee na ene acre em 2.85 2.70 2.85 2.74 7A to Tt ON ee oe oe OV ERAPE 2 foie sans /-b issn nian 2.59 2.85 2.74 2.87 | 2.71 ? Allen’s Commercial Organic Analysis, Vol. III, Pt. II, p. 485. 24 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Caffein determinations on tea and coffee—Continued COFFEE. FULLER GORTER STAHLSCHMIDT METHOD METHOD MODIFIED METHOD HH Hansons.2e cee see eee 1.28 1.36 1.11 1.41 1.21 11.60 1.28 STs) 1.23 Ae Gi Fullerc3):.j..c..caheeeiiot eee 1.12 1206) il 28a 5itee| ee seers lee eee HM. Js Wichmann. s. 300k ce eeeee ete eee | cena oar levees ere 1.40 Phy) 1.41 12 A [a © Be ©) E:hd <: Pea icc coca badecstoonlloapetdcd Wooo am ac 1.31 1.12 1.34 1.16 1.29 1.14 J. M. Bartletts oj: oticwediowc anton aeeiee Bee Greece oialn| ase ieernet 1.39 1.21 1.40 1.34 1.32) 4 eee 1.31 15:25 ACVEr Age s(ya-Wer itn elo trea 1.20 ie eal he 1.36 1.20 1 Omitted from average. One analyst, H. J. Wichmann, has commented on the method, stating that the caffein from the tea seemed sufficiently pure to weigh directly, but that from the coffee was contaminated with some yellow ma- terial. The method as sent out was the same for coffee as for tea, but making the solution alkaline with ammonia before shaking out with chloroform gives a much whiter residue in the case of coffee. The results obtained are very satisfactory when we consider the ma- terial, and more concordant than most results obtained in previous years by other methods, particularly when calculated from the nitrogen con- tent. However, most of the results obtained by direct weighing are suffi- ciently accurate for the purpose of estimating purity of the materials from the caffein content. So few analysts have taken part in the work and sc few varieties of teas and coffees have been analyzed that your referee does not feel warranted in yet recommending the method as an official one, but considers it highly preferable to the present provisional methods given in U.S. Bureau of Chemistry Bulletin 107 (revised), which, as far as he can learn, are not being used by anyone for the determination of caffein in these products. It is, therefore, reeommended— (1) That the Stahlschmidt method as modified in this paper be adopted as a provisional method for the determination of caffein in tea and coffee. (2) That the method be further tried on a greater variety of teas and coffees, anticipating its adoption as an official method. (3) That the referee for next year study methods for determining tannin in tea and coffee. 1917] GNADINGER: SACCHARIN IN FOODS 25 DETERMINATION OF SACCHARIN IN FOODS. By C. B. Gnapincer (Bureau of Chemistry Food and Drug Inspection Laboratory, Chicago, III.) The provisional method of the association’ for the determination of saccharin in foods is, briefly, to macerate the solid or semisolid material with dilute alkali, centrifuge, acidify the alkaline solution, extract the saccharin with ether, evaporate the ether, and determine in the ether ex- tract the sulphur from which the weight of saccharin is calculated. This method is not applicable to foods containing an appreciable amount of ground mustard, because the ether-soluble sulphur compounds present in mustard are extracted, in part, with the saccharin and are not separated from it by washing the ether extracts with dilute alkali. Difficulty in obtaining a clear solution for extraction, precipitation on acidifying the alkaline solution, and formation of very troublesome emulsions also ren- der this method objectionable. _ The experiments here described were made for the purpose of develop- ing a method free from these objections. Incidentally some of the prop- erties of saccharin were investigated. SELECTION OF SOLVENT. For the purpose of selecting the solvent most suitable for the extrac- tion of saccharin, its solubility in various solvents was determined, and experiments were made to establish the number of shake-outs necessary for its extraction from aqueous and acid solutions. The saccharin used had the following analysis: Moisture (loss on drying over sulphuric acid)....................... None Piher-insolublemmatteri(per Cent). oss. de. cccc. ccanscceeesceeee 0.32 Benzoic sulphinid, by Reid’s method? (per cent).................... 96.20 Nitrogen\presentasiammonium Salts. 2... << scene die ceccceocieeee None Total acidity calculated as saccharin (per cent)..................+. 98.03 Saccharin calculated from sulphur determined by fusion (per cent).. 98.09 USHG IA END OL Cao yxter terete ssieis a=) a sioteistater keystones raising csi sec.s se oloeeeere 224 In determining the solubilities, a slight excess of saccharin was added to a portion of the filtered solvent in a glass-stoppered flask and allowed to stand several days at 27° to 30°C., with occasional shaking. The flask was then placed in a water bath at 25°C. for at least one hour and shaken frequently. Part of the solution at 25°C. was transferred to a weight burette by means of a pipette, closed with a piece of filter paper, and weighed from the burette into a tared dish. The solvent was evaporated spontaneously before a fan, and the residue dried to constant weight in 1U. 8. Bur. Chem. Bul. 107 (rev.), p. 182. 2 Allen’s Commercial Organic Analysis, 4th ed., vol. 3, p. 434. 26 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 a desiceator. Similarly, the percentage of solids in the solvent was determined and a corresponding correction made. From the results thus obtained the weight of saccharin dissolved by 100 grams of solvent at 25°C. was calculated. TABLE 1. Solubilities of saccharin. SACCHARIN SACCHARIN DISSOLVED DISSOLVED SOLVENT é Br SOLVENT ens SOLVENT AT SOLVENT AT 25°C. 25°C. grams grams Acetone sis: )-se-eescreeioeic eto 14¥S 7p Chloroform perme. a peeeeiee 0.259 Methyl acetate................ 8309) || (Benzolia.tr-areecsea iene sinenieies 0.129 Hthyliacetatesn-casst ee ee 4:50) oluol eie.8. eden esas meee 0.113 Ethyl alcohol (99.5%)......... St SAM leXsy) ols a3 he uate reer tee 0.097 Amyliacetates:-)cocemeee ose 1.69 | Carbon tetrachlorid.......... 0.0067 Ether (‘‘over sodium’’) ....... 1256) )|\Carbonibisulphidememeeees eee 0.0064 ‘Amv yii Valcoholeeien ert secs 1.22 | Petroleum ether, boiling point Water (distilled).............. 0.400 Eiht OMe Baa ddoasaaoedac 0.0009 Those solvents in which saccharin is much less soluble than in water need not be considered further. Of those remaining, acetone and ethyl alcohol are miscible with water, and methyl acetate is largely soluble in water. Amyl acetate and amyl alcohol have the disadvantage of high boiling points. Ethyl acetate, ether, and chloroform were next compared as to the num- ber of extractions necessary for the removal of saccharin from aqueous and acid solutions. The following solutions were prepared: Saccharin, 1 gram per liter and 2 grams per liter; 2 N HCl; N/5 HCl; 2 N acetic acid; N/5 acetic acid; 2 N acetic acid plus N/5 HCl. The ether and ethyl acetate used were washed three times with water and filtered; the chlo- roform was filtered. One hundred cubic centimeters of solution containing 100 mg. of sac- charin were pipetted into an 8 oz. separatory funnel and 50 cc. of the immiscible solvent added from a pipette. The separatory was shaken vigorously for two minutes, allowed to stand at least fifteen minutes, and the immiscible solvent drawn off into a tared dish. The extraction was repeated twice, and the three portions of solvent, in separate tared dishes, were evaporated before a fan. The dishes were then dried to constant weight in a desiccator. A blank was run, using 100 ce. of distilled water and 50 cc. of solvent, and a correction made for the solids thus found in the solvent. The experiment was repeated, extracting 50 cc. of saccharin solution containing 100 mg. saccharin and 50 cc. 2 N HCl or 2 N acetic acid, ete., with 50 cc. of solvent. Blanks were run, using 50 cc. of distilled water instead of the saccharin solution. In some cases the extracts were ti- 1917] GNADINGER: SACCHARIN IN FOODS 27 trated with N/100 NaOH (phenolphthalein) as well as weighed. The agreement between gravimetric and volumetric results was excellent. All extractions were made at room temperature, 27°-30°C. TABLE 2. Extraction of saccharin with chloroform, ether, and ethyl acetate from aqueous and acid solutions. AQUEOUS SOLUTION SACCHARIN EXTRACTED (100 cc. EXTRACTED) = TOTAL IMMISCIBLE SOLVENT (otlcoSUBED) Sac- ree | essen) || gies | ee aeed ee, a ana (ae || mg mg. mg. mg. per cent Chloroform...... TC OEW OBS seetes cere aoe rere if Ae on re aoe ogee) 100 | Acetic, N/10............ HH MAN eels ea Bia atlas 100 | Acetic, N/l............ yee arse cdi eae Diet tO MECI NYT 2 a Feah RCaeae rees beetee Bone es LOO METOL INA 2 ee. oe TRE eae eee ee. a1 | (ea, aaa 30.7 | 20.5| 120) 63.2 1D inte oeeere LOO |WAcetic: IN/ 1s sex. aces oe aad ae sage Le Daaiee << 100 | Acetic, N/1;HCI,N/10.) 74-5] 20-5] 2-2) 9.2 lbs eee TOO MELO USN OG ce ces cietocre ot ns ioe ee eae Dee. THEE NA 2 sis| 148| 31) 997 Ethyl acetate....| 100 | HCI, N/10............. ae a6 pe hae Either ethyl acetate or ethermay be used forthe extraction of saccharin, two extractions being necessary with the former and three with the lat- ter, when one volume of solvent is used to extract two volumes of hydro- chloric-acid solution. Saccharin is more readily extracted from hydro- chloric-acid solution than from acetic-acid solution. Chloroform is not a suitable solvent. DETERMINATION OF SACCHARIN IN MUSTARD PRODUCTS. In detecting saccharin by the method of Bianchi and di Nola,! the solu- tion acidified with acetic acid is clarified with lead acetate, made acid with sulphuric acid, and extracted. This method, slightly modified, applied to prepared mustard removes very little of the interfering sulphur com- pounds. If, however, the solvent used for the extraction be evaporated and the residue treated with petroleum ether, most of these compounds 1 Allen’s Commercial Organic Analysis, 4 ed., vol. 3, p. 482. 28 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMIsTs [Vol. III, No. 1 are removed. The petroleum-ether extract contains a very pungent, oily, sulphur-bearing substance. By further treatment of the residue with bromin, practically all of the interfering substances are eliminated while the saccharin remains unchanged. The lead-acetate clarification yields a solution which can be readily extracted with the formation of little or no emulsion. The following procedure is based on the method mentioned above and on the preceding solubility experiments: METHOD FOR DETERMINATION OF SACCHARIN IN MUSTARD PRODUCTS. Transfer 50 to 75 grams of the material (ground in a meat grinder, if necessary) to a 250 cc. volumetric flask with nearly boiling water, diluting to about 200 ec.; let stand 2 hours, shaking occasionally. Add 5 ee. glacial acetic acid, mix thor- oughly and add a slight excess of 20% normal lead acetate solution. Make to mark with cold water and Jet stand 20 minutes. Centrifuge and pour the supernatant liquid through a folded filter. Transfer 150 cc. of filtrate to a separatory funnel, add 15 ec. concentrated HCl and extract three times with 80 ce. portions of ether, shaking the separatory for two minutes each time. Wash the combined ether ex- tracts once with 5 ec. water and transfer the ether to a 250 cc. beaker. Add about 10 grams washed sea sand and evaporate the ether before a fan or air blast. Dis- tribute the sand on the walls of the beaker with a stirring rod and continue the spon- taneous evaporation until quite dry. Add 25 ce. petroleum ether (boiling point, 30° to 65°C.) and rub thoroughly with a “‘policeman.’’ Decant through a dry 7 em. quantitative filter paper and repeat the washing twice, using 25 ec. petroleum ether each time. Reject the petroleum ether washings and return the filter paper to the beaker containing the sand. Wash the residue on the sand with hot water and fil- ter into a separatory funnel, collecting about 75 ce. of filtrate. Cool, add7 to8 ce. concentrated HCl and a distinct excess of bromin water. Let stand 5 minutes and destroy the excess of bromin with sodium nitrite solution, avoiding a large excess of the latter. Extract the acid solution three times with 50 ce. portions of ether and wash the combined ether extracts once with 5 ce. water. Evaporate the ether spontaneously and determine the sulphur in the residue by fusion with so- dium peroxid or a mixture of six parts sodium carbonate and one part potassium nitrate. Conduct the fusion in a nickel crucible. Weight of BaSO, multiplied by 0.7844 gives the weight of saccharin. To the weight thus found add 0.5 mg. to cor- rect for the saccharin dissolved by the petroleum ether. A blank should be run to determine sulphur in the fusion mixture. The determination of sulphur in saccharin by the fusion method re- quires very careful manipulation to prevent loss of sulphur because of imperfect oxidation. Experiments were made to determine the possi- bility of converting the sulphur to sulphate by electrolyzing solutions of saccharin. ELECTROLYTIC DETERMINATION OF SULPHUR IN SACCHARIN. If a solution of saccharin in sodium hydroxid be electrolyzed under proper conditions, the sulphur in the saccharin will be converted quanti- 1917) GNADINGER: SACCHARIN IN FOODS tatively into sulphate, which can then be determined as BaSOs,. acid or potassium hydroxid can be substituted for sodium hydroxid, but nitrates and potassium salts are occluded by the BaSOs precipitate, while sodium salts do not interfere.! 29 Nitric The results obtained by electrolyzing solutions of saccharin under dif- ferent conditions are given in Table 3. The ordinary, 110-volt, direct no. ood oan kr wry ee Y ao fF WO NY & © 16 TABLE 3. Determination of sulphur in saccharin by electrolysis. MATERIAL OF ELECTRODES AREA oF AMPERES TIME OF BLEC- TROLYSIS SOLUTION anopE | (ASE (100 ce.) sg. em, | 2g.em. re. OMe ace ce ei 3.8] 117.0] 2.8-2.9| 3 |NaOH, Ny1 RON retopinids teatiere oe 117.0} 3.4] 2.8-2.9| 3 |NaOH, N/1 ISI aL os Seana 1.2] 117.0] 2.7-2.8| 3 |NaOH, N/1 Nickelk ev c.aus e's a 117.0] 1.1] 2.7-2.8| 3 |INaOH, N/1 Platinized platinum.| 117.0| 1.0) 2.7-2.9}| 3 |NaOH, N/1 Smooth platinum.... 0.9|117.0| 2.8-2.9 | 3 |NaOH, N/1 Smooth platinum.... 0.9] 1.0] 2.8-2.9| 3 INaOH, N/1 Smooth platinum....| 125.0} 92.0] 2.8-2.9| 3 |NaOH, N/1 Smooth platinum....}| 117.0} 1.0| 2.7-2.8| 3 |NaOH, N/1 Smooth platinum....| 117.0} 1.0) 0.7-1.0| 3 |Aqueous Smooth platinum....| 117.0} 1.0} 2.4-2.6| 3 |NaOH, N/10 Smooth platinum....| 117.0} 1.0| 2.7-2.8| 3 |NaOH, N/2 Smooth platinum....| 117.0} 1.0) 2.7-2.8| 3 |NaOH, 2N Smooth platinum....| 117.0} 1.0) 0.45 | 3 |NaOH, N/1 Smooth platinum....| 117.0! 1.0] 1.5-1.6|} 3 |NaOH, N/1 Smooth platinum....| 117.0} 1.0) 2.7-2.9| 6 |NaOH, N/1 Smooth platinum....| 117.0] 1.0) 2.7-2.9| 6 |NaOH, N/1 Smooth platinum....} 117.0] 1.0] 2.7-2.9| 6 |NaOH, N/1 Smooth platinum....| 117.0} 1.0) 2.7-2.9| 6 |KOH, N/1 Smooth platinum. . pelt LEO 1.0| 2.7-2.9} 6 |HNO;, N/1 Smooth platinum....| 117.0, 1.0] 2:7-2.8| 6 |NaOH, N/I fe er Sees 1 Treadwell. Analytical Chemistry 2: 368. SACCHARIN SACCHARIN RECOVERED 3 eee esr om renee SE — nn ee ee eeeee—e eee es ae rr rr, _e=eren— X=~=——_— Ee Oe CONMICOCONIMD EH Dr MWNOCNAMNOHNHWWMNOWNPOARQRNUNOWDWUINRPODHOCOPMAOANCUNIA — — SO alae O11 GH Go © 00 00 CO CD ND to CO 0000 0 30 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 current was used, the strength of the current being regulated by lamps of various sizes connected in parallel; an ammeter was also placed in the circuit. Cylinders and wires were used as electrodes. Twobinding posts, screwed to a wooden bar clamped ona ringstand, supported the electrodes. Experiments 1 to 9, inclusive, show the effect of using electrodes of different metals; in 6, 7, 8, and 9 the size of the electrodes was varied. Solu- tions containing different concentrations of NaOH were electrolyzed in experiments 9, 10, 11, 12, and 18, while in 9, 14, and 15 the strength of the current was the varying factor. A comparison shows that smooth: platinum electrodes give better results than iron, nickel, or platinized platinum electrodes. The relative size of the electrodes is not of primary importance, but it is better to use a large anode and a small cathode. The concentration of the NaOH solu- tion should be about normal. The current strength should be 2.7 to 3.0 amperes. Accordingly, different amounts of saccharin were electrolyzed for 6 hours with a current strength of 2.7 to 2.9 amperes, using smooth platinum electrodes, the saccharin being dissolved in 100 ce. of N/1 NaOH, N/1 KOH, or N/1HNO;. Solution containing saccharin and benzoic acid or saccharin and salicylic acid were also electrolyzed. All reagents were tested for sulphur. Results are shown in experiments 16 to 22. In determining saccharin in foods, dissolve the residue obtained from evaporating the ether extract in hot water and filter. Collect about 75 ec. filtrate and washings in a beaker and to this solution add 25 ce. of 16% NaOH solution. Continue the electrolysis for six hours, using smooth platinum electrodes (area of anode, approximately 120 sq. em.; area of cathode, approximately 1 sq. cm.) and a current strength of 2.7 to 2.9 amperes. Filter the electrolyzed solution, make faintly acid with HCl, and precipitate with BaCl.. Weight of BaSO, multiplied by 0.7844 gives weight of saccharin. Run a blank to determine the sulphur in 25 ec. of the 16% NaOH solution. The blank need not be electrolyzed. Results obtained on mustard products by the method described, as well as by the lead clarification method and the provisional association method, are given in Table 4. Samples A, B, and C were different brands of pre- pared mustard. Sulphur was determined by fusion with NazCOs;, and KNOs, and by the electrolytic method. APPLICATION OF THE LEAD ACETATE METHOD TO VARIOUS FOODS. In applying the lead acetate clarification to foods containing no mus- tard the procedure described can be materially shortened as the petro- leum ether extraction, bromine oxidation and second extraction with ether are unnecessary. The following method differs slightly from that of Bianchi and di Nolla previously mentioned. 1917] GNADINGER: SACCHARIN IN FOODS M 31 TABLE 4. Determination of saccharin in mustard. SULPHUR FOUND CALCULATED AS SACCHARIN SAC- Sample A Sample B Sample C METHOD CHARIN | J} | ___ PRES- By By By aN By elec- By elec- By elec- fusion tro- | fusion | tro- |fusion.| tro- lysis lysis lysis per cent|per cent|per cent|per cent|per cent|per cent|per cent Brovaisionalin sa -\scisiss os isicis wei i itiasie ays ae Leiendwas aes oS 0.017-0 017 fEamarind paste (commercial sample)_.........- 2 t0.02--2-s0c-2cesceeese. 0.030-0 .033 Vinegar from mixed pickles which showed no formic acid by the Fincke HEE RGR MEET ee ee settee eee sae see ooh swede ste 0.007—-0.057-0 .040 Comparing these results with those obtained during the past two years with the Finéke method, it is evident that the latter is to be preferred for accuracy and reliability. It would appear from the results obtained by Mr. Lyman and the referee upon pure samples, and also by the reports of the collaborators, as given in Table 3, that at times some impurity passes over into the distillate which, during the decomposition of the dry residue with concentrated sulphuric acid, tends to increase the volume of gas formed beyond that due to the formic acid alone. Comments received from collaborators were as follows: M. G. Wolf believes that the Wegner method is too time consuming and re- quires too much attention. E. R. Lyman prefers the Fincke method for accuracy and reliability. R. W. Hilts has been unable to report upon the samples sent out by the referee owing to pressure of other work, but, having had experience in the use of both the Fincke and the Wegner methods, writes as follows: “‘There is no denying the fact that this method (Wegner’s) is time consuming and that it requires the very close attention of the analyst during decomposition of the sodium formate. I will be interested to see what the comments of the collaborators will be. The method is not to be compared to Fincke’s method for routine use. It is, how- ever, the best confirmatory method that I know and combines a qualitative value with the quantitative determination. 1 believe that lactic acid is practically the 38 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 , only common acid at all volatile which would yield carbon monoxid under the conditions outlined; and in view of the slight volatility of lactic acid, this objec- tion is not very important. I believe that it would be desirable to have this method available for use as an optional method for confirmation in contested cases. If the collaborators’ results are very unsatisfactory, of course this opinion might be somewhat modified.” Taking into consideration the time, attention and labor involved in the Wegner determination and also the results reported by the collabora- tors upon the samples sent out, it is considered inadvisable to recommend its adoption as an official method. The referee is of the opinion, however, that the method is useful for confirmatory purposes in cases when the amount of formic acid to be determined is not extremely small, due regard being paid to the apparent limits of error for the method. It is recommended that the Fincke method adopted provisionally last year be now made official. The referee wishes to express his acknowledgments to Messrs. A. L. Burns, L. D. Elliott, E. R. Lyman, L. Katz, W. L. Scovill, and M. G. Wolf for their assistance in this work. SACCHARIN. By A. F. SkeKer anp M. G. Wo tr. An investigation has been made of several methods that have been pro- posed for the determination of saccharin. These have all been found to provide for the extraction of the saccharin from solutions of the substance in which it exists by means of immiscible solvents and subsequent de- terminations in the residue remaining after evaporation of the solvent. This residue has been weighed directly,! converted into the silver salt and weighed,? converted into salicylic acid and determined as such,’ or by determination of the sulphur in the residue. Testoni also hydrolizes this residue by heating under pressure with 1:1 hydrochloric acid and determines the ammonia formed by distillation into standard acid. Since all the proposed methods rely upon a complete extraction of the saccharin by shaking an aqueous solution containing it with an immis- cible solvent, it was decided first to ascertain which of these is most sat- isfactory for the purpose in view. Ether and mixtures of ether and benzene and ether and petroleum ether have been used by various work- ers. Marden® found that chloroform is inefficient, but that ether ex- ?Tortelli and Piazza, Z. Nahr.-Genussm., 20: 489; Karas, ibid., 25: 559; Testoni, ibid., 18: 577; Possetto and Issoglia, Giorn. farm. chim. 61: 5. 2'Testoni, loc. cit. *Carlinfanti and Marzocchi, Boll. chim. farm., 47: 599. *Van den Diesen, Apoth. Ztg., 22: 230; Testoni, loc. cit., U. 8S. Bur. Chem. Bul. 107 (rev.), p. 183. 6 J. Ind. Eng. Chem. (1914), 6: 315. 9 1917) SEEKER: PRESERVATIVES 39 tracts the saccharin quantitatively with relatively small amounts of the solvent if the water layer is sufficiently acid (5 ec. concentrated hydro- chloric acid to 100 ce. of solution); he also recommends amyl acetate as a solvent. As a preliminary step, pure saccharin was prepared by acidifying a strong aqueous solution of the commercial sodium salt, washing the pre- cipitated saccharin with water, and recrystallizing from hot water. After powdering and drying the product over sulphuric acid, it was found to yield no residue on ignition, and the sulphur, nitrogen, and ammonia (formed by acid hydrolysis) were found to agree well with the theoretical for pure saccharin. As a result of a series of trials upon pure water solutions and upon aqueous extracts of jams containing known amounts of saccharin, it was found that ether or a mixture of ether and petroleum ether (equal parts) extract the saccharin quantitatively in three or four extractions with relatively small amounts of solvent, particularly if the aqueous layer be nearly saturated with salt. Benzene, or a mixture of benzene and ether, -under the same conditions give unsatisfactory yields. In all cases when ether alone was used notable amounts of salt were retained in the solvent, and this could not be washed out without at the same time removing part of the saccharin. When aqueous extracts of food substances like fruit pulp or jam were used all the solvents retained so much impurity that the saccharin could not be weighed as such, but was estimated by fusion of the residue with sodium and potassium carbonates and de- termination of the sulphur as barium sulphate. Repeated washing of the saccharin solution in the volatile solvent previous to evaporation failed to remove the impurities, and the preliminary treatment of the food material with lime and alcohol-salt solution previous to extraction as proposed by Tortelli and Piazza not only failed to remove this difficulty but also yielded low recovery of the saccharin as determined by alkaline fusion of the residue. Determinations were then made as follows: 50 grams of the matcrial were weighed into a porcelain dish and converted into a thin paste by thorough mixing with a small amount of saturated salt solution. The mixture was then introduced into a 250 cc. volumetric flask and the dish rinsed with salt solution, the rinsings being added to the contents of the flask. About 5 cc. of milk of lime was then added and the volume made up to 250 ce. with saturated salt solution. After vigorous shaking the mixture was filtered through a dry filter and an aliquot portion of the filtrate, usually about 150 ce., mtroduced into a separatory funnel, acidi- fied with hydrochloric acid, and extracted with three successive 50 ce. portions of ether. Each of the ether extracts was washed in succession with two portions of 3 cc. each of saturated salt solution, the combined 40 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 ether extracts filtered through a dry filter, and evaporated to dryness on a steam bath. The residue was then transferred to a platinum cru- cible by means of a little ether, again evaporated to dryness, the residue moistened with a few drops of sodium carbonate solution and fused with 3 grams of a mixture of equal parts of sodium and potassium car- bonates. The melt was dissolved in water, acidified with hydrochloric acid, and the sulphur determined as barium sulphate. Proceeding in this way, the following results were obtained: TABLE 1. Saccharin in food products. SACCHARIN SUBSTANCE a RECOVERY Added Found mgs. mgs. per cent Orange marmalades neces ok er -teleetes etatelee clo 50 45.7 91.4 Apricotipulp eee eee eee een eee eee eee 25 23.3 93.2 Rhbubarbisauce ese ease steocieseraeee elses 55 51-1 93.0 Doe 2 ERA ERE Sea. soretereprroecr teeta 70 63.5 91.0 In order to duplicate as nearly as possible the type of sample repre- sented by substances of the nature of ice-cream cones, four loaves of bread were prepared from the same flour, each approximately 1 pound in weight when baked, three loaves containing, respectively, 100, 200, and 400 mgs. of saccharin, the fourth loaf, containing no saccharin, serving as a blank. The finely powdered saccharin was intimately mixed with the flour before making the dough, and yeast alone was used as a leavening agent for the latter. The bread, after baking, was cut into large pieces and dried for a short time in a water oven. It was then ground to a moder- ately fine powder in a hand mill, no appreciable loss of material taking place in this process, the product obtained in this way being regarded as containing all the saccharin originally added to the flour, and its weight being used in calculating the recovery of saccharin from aliquot parts. The powder, after weighing, was preserved in tightly stoppered bottles. The method described above was used for the extraction of the sac- charin, except that it was found necessary to correct the volume of the alkaline salt solution for the space occupied by the bread, and the mixture of bread, milk of lime, and salt was allowed to stand several hours before filtering, in an attempt to secure complete solution of the saccharin. It was found that the bread containing no saccharin gave a consider- able blank in the sulphur determination upon the residue from the ether extract. This was in every case deducted from the determinations con- ducted upon the samples containing saccharin. The results were as follows: 1917] SEEKER: PRESERVATIVES 41 TABLE 2. Saccharin in bread. SACCHARIN PRESENT SACCHARIN FOUND RECOVERY mg. per loaf mg. per loaf per cent 100 62.3 62.3 200 160.9 80.5 400 264.0 66.0 As might be expected, there appears to be some difficulty in extracting the saccharin completely from the bread, and further work is in progress to secure a better method for accomplishing this. Concerning the means for determining the saccharin in the impure residues obtained after removal of the volatile solvents used for extrac- tion, an estimation of the sulphur seems to have found the most extended use. Direct fusion of weighed amounts of pure saccharin as in the method given above ytelded results agreeing closely (98-100%) with the theo- retical, and there appears to be no reason to doubt its accuracy when applied to the residues in question, provided ether-soluble sulphur com- pounds, like mustard oil, are absent (see pp. 25-32). Testoni proposes to weigh the saccharin as silver saccharinate. In order to test this procedure, 0.114 gram of pure saccharin was dissolved in 10 cc. of hot water, 10 ce. of alcohol added, and the mixture cooled. Aiter adding 10 cc. of a saturated solution of silver nitrate in alcohol, the mixture was allowed to stand overnight, filtered upon a tared Gooch, the precipitate washed with 15 cc. of absolute alcohol, dried in a water oven, cooled, and weighed. The weight of the silver salt represented 90% of the saccharin used. The experiment was repeated, employing a few drops of a saturated aqueous solution of silver nitrate as a precipitant, and adding 50 mgs. of sodium acetate to overcome the solvent effect of the free nitric acid liberated by the double decomposition between the silver nitrate and the saccharin. The silver salt in this case represented 97% of the saccharin used. : Testoni also proposes the determination of the saccharin in the impure residues by means of the ammonia formed by acid. hydroly- sis, distilling the hydrolized mixture for this purpose. Since the amount of ammonia formed in an actual determination is very small, this idea has been applied in modified form by weighing as ammonium chlorplatinate. The method of hydrolization has also been modified, and instead of heat- ing under pressure as proposed by Testoni, it has been found that by boil- ing for three-fourths of an hour under a reflux condenser with a small amount of dilute hydrochloric acid (1 part concentrated acid and 9 parts water) complete hydrolysis takes place. After hydrolysis, the acid mix- ture is treated with a small amount of platinic chlorid solution (the 42 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 weight of dry platinic chlorid used to be twice the expected weight of saccharin), evaporated to dryness on a steam bath, the residue taken up in alcohol, filtered on a tared Gooch, the precipitate washed with alcohol, dried at 130°C., and weighed as ammonium chlorplatinate. Working in this way with pure saccharin the recoveries were 94, 100, and 100% in three determinations. Accurate results have not as yet been obtained by applying either the silver salt or the ammonium chlorplatinate methods upon the impure saccharin obtained from food samples, owing to the disturbing effects of the impurities. A combination of the two has, however, been found to be fairly successful. By hydrolizing the silver salt with hydrochloric acid as described above, filtering, and converting the ammonium chlorid in the filtrate into ammonium chlorplatinate, the following results were obtained: TABLE 3. Saccharin in food products. SACCHARIN SUBSTANCE RECOVERY Added Found mg. mg. per cent Orange marmalade................. 55 57.6 104.7 Chow chow:2ssceecrce come cnr 30 30.6 102.0 It is proposed during the next year to formulate the Testoni method, or a modification of it, so that it may be used as a general method for the determination of saccharin in foods, and if possible to submit the pro- cedure to collaborators for trial. It is hoped that the method may be so modified that the saccharin may be weighed as such previous to its determination as one of its decomposition products. It is recommended that this work be continued. Other references upon this subject are: Genth. Am. J. Pharm., 81: 536. Bianchi and di Nola. Chem. Zentr. (1908), 2: 2039. Flamand. Bull. soe. chim. belg., 26: 477. Ledent. Ann. chim. anal., 18: 314. Parmeggiani. Z. oesterr. Apoth.-Ver., 46: 179. Condelli. Boll. chim. farm., 52: 639. 1917] TREUTHARDT: HEAVY METALS IN FOODS 43 REPORT ON HEAVY METALS IN FOODS. By E. L. P. Treurnarpt (Bureau of Chemistry, Washington, D. C.), Associate Referee. Owing to the large field covered by this subject, it was found impossible to undertake all of the work recommended in the report of last year. The work was confined to testing those methods for the determination of arsenic and tin which were recommended for further study and which, it was hoped, could be recommended for adoption by the association. In accordance with the plans of last year, the study of methods for the determination of lead in baking powder and baking-powder materials was conducted by Dr. H. E. Patten, associate referee on baking powder, and is made the subject of a separate report. ARSENIC. The procedure for ‘‘ Method I” in the report of last year was rewritten as given below so that no references would be necessary to published directions. The 1:8 sulphuric acid strength in the generators was re- tained, as it is considered far more satisfactory than the stronger acid, as it is less liable to form hydrogen sulphid during reduction. METHOD FOR THE DETERMINATION OF ARSENIC, Adapted from method of C. R. Smith (U.S. Bur. Chem. Cire. 102). (See Sanger and Black. J. Soc. Chem. Ind., 1907, 26 : 1115). REAGENTS. Concentrated nitric acid and concentrated sulphuric acid.—Should be arsenic-free. Dilute sulphuric acid.—1: 4. Zinc.—Arsenic-free stick zine broken in pieces 1 inch long. Lead acetate paper.—Heavy filter paper soaked in 20% lead acetate solution, dried and cut into pieces about 4.5 by 16 em. Lead acetate cotton.—Absorbent cotton soaked in 5% lead acetate solution. Mercuric bromid paper.—Cut heavy, close-textured drafting paper (Whatman’s cold-pressed if possible) into strips exactly 2.5 mm. wide and about 12 em. long. Soak an hour in 5% solution of mercuric bromid in 95% alcohol and dry on glass rods. Cut off the ends of the strips before using. Potassium todid solution.—20%. Stannous chlorid solution.—40 grams of crystals made up to 100 cc. with concen- trated hydrochloric acid. Standard arsenic solution.—Dissolve 1 gram arsenious oxid in 25 cc. of 20% sodium hydroxid, neutralize with dilute sulphuric acid, using litmus paper, add 10 ce. con- centrated sulphuric acid, and dilute to 1 liter with recently boiled distilled water. One cubic centimeter of this solution = 1 mg. As.O3. Twenty cubic centimeters of this solution is diluted to 1 liter; 50 ec. of the dilute solution is made up to 1 liter; 1 cc. of this last solution = 0.001 mg. As20;. This solution is used to make the standards. The dilute solutions should be made up freshly when required. 44 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS Vol. III, No. 1 APPARATUS. The generator bottle is a 2-ounce wide-mouth jar, connected by means of a rub- ber stopper to a glass tube 1 cm. by 6 cm.. containing a piece of lead acetate paper rolled into a cylinder. This is connected with a similar tube loosely filled with cotton moistened with 5% lead acetate solution. The cotton should be uniformly moist in all tubes. The second tube is connected with a capillary tube 3 mm. in internal diameter and 12 cm. in length, which contains the mercuric bromid paper. All connections are made with rubber stoppers, from which white coating should be removed. DETERMINATION. Weigh 25 grams of sample into a porcelain casserole, add 10 cc. arsenic-free nitric acid and cover by setting a watch glass inside the rim, convex side upward. Heat until vigorous action is over, cool, and add 10! cc. arsenic-free sulphuric acid. Heat on a wire gauze over a flame until the mixture turns dark brown or black, then add more nitric acid in 10! ec. portions, heating between each addition until the liquid remains colorless or yellow, even after the evolution of SO;fumes. To remove com- pletely all nitric or nitrous acids, evaporate to 5 cc.; and if on addition of water to the cooled acid nitrogen peroxid fumes are evolved, a second evaporation to white fumes is necessary. Dilute the acid solution, transfer to a 100 cc. flask and rinse out the casserole with water. Cool and make up to 100 cc. with water. Introduce 20 cc. of this solution into a 2 oz. generator bottle, add 20 cc. 1: 4 sulphuric acid and 4 cc. potassium iodid solution. Heat to about 90°C., add three drops stannous chlorid solution and continue heating for ten minutes. Cool the bottles in a pan containing water and ice. When cold, add about 15 grams? stick zine in several pieces and connect with the tubes. Keep the bottles in ice water for fifteen minutes; take out and allow to run for one hour longer. Then remove the paper and compare with standard stains. Run a blank test with reagents alone. STANDARDS. Measure out portions of the dilute standard solution containing from 0.001 to 0.050 mg. As2Os3, and add proper quantities of water and sulphuric acid so that the generator bottle contains 40 cc. solution of 1: 8 sulphuric acid strength. Add potassium iodid and stannous chlorid and proceed as directed under “Determination.’’ Heating, cooling, and all other conditions should be the same for standards as for determinations. COLLABORATIVE WORK. Two samples were prepared for testing the methods. One was a 25% sugar solution containing 1.4 mg. As,O; per kilogram and the other was a 7.5% gelatin solution containing 12.0 mg. AseO3 per kilogram. Results were reported from eleven collaborators as follows: ‘Experience in this year’s work has shown that 10 ce. nitric acid is frequently excessive. Portions of 3 to 4 cc. are ordinarily sufficient. See comment by Black, p: 45... * This may not be enough zine. In determination made by the referee, 30 to 40 grams were used. 1917) TREUTHARDT: HEAVY METALS IN FOODS 45 Codperative results on arsenic. [Reported in milligrams As:Os per kilogram.} SAMPLE 1 SAMPLE 2 ANALYST Sugar solution contain-| Gelatin solution con- ing 1.4 mg. As2Os taining 12.0 mg. As:Os per kilogram per kilogram 1.9 14.3 <1 te { i a8 9.0 \/> 1b. ine eeseoebeesseoedsacuensnpepeoods 123 { 98 ASSESSOR ODED AALS 10.0 1.8 12.0 PrpPBE NIC OGE seco ves vis oic.,c ete ss oo mains 2.0 16.0 1.6 15.0 ee ee *20.0 . b 10.2 mltcldstowtesereerere ctiic ees sores onatng os \ 08 101 IEMA Pore nia saeco sje ec(ein els esiv se seine *4.0 *20.0 rH ETO Gee ris ee etnies 2 aa inys aisle starevere areioe ess 222 12.8 HS PARAL PEN cane esies ce oe'eis.d clevssioesle t Saiclee *0.6 *5.8 OMG OLCELE eer aye aro sre siecle asin ose Saree ais 20 *5.0 > * CVPR LEAINIOD Soo KOaiats e siecrele siseis aseine oie { ae a | 1.8 10.0 ¢ 1.8 12.0 Pree Eee Preu chard bocce sticicacce cctecit +s ss a 12:8 l i.4 13.2 EIRP APE VV NILIATYe Pennimangeareer nr ciace are ji 19.8 ee 15.0 ee a TE | Samet eigen) fi Ameer PAN Seige OD Oda: ll. He Ri: Smithy eee orcs atom. a aoa) | Deee ae 116 GP We Trainor’ eer scene eee 22.6 20.9 13.6 14.1 } 20.6 20.8 13.8 He Lt Pe Dreuthardton: were 23.2 22.9 14.2 14.2 i 21.4 220i. 14.1 14.1 A A. Whitman Secsee. acorn PPA 20.2 12.7 12.0 Highest).jt2-)ee eee cee ee oe 26.0 23.4 15.5 15.0 LOWeS ti aia saastete tigers Sacre eee yeaeae 19.8 20.1 10.3 9.5 AVCTAZO. 0.222 Selys mis ats poccice semelrminne 23.0 22.2 13.6 13.0 2 Tritrated with N/20 iodate solution. Two of the collaborators considered their results by the gravimetric method too unsatisfactory to report. One collaborator was unable to report results by the volumetric method for the same reason. COMMENTS BY COLLABORATORS. L. D. Elliott: The reduction of the tin in the volumetric method requires prac- tice in manipulation before reliable results can be obtained. In my duplicate of each sample I reduced the solution with antimony and connected the flask, after boiling the contents two minutes, by means of a tube to a solution of sodium bicar- bonate. H. A. Whitman: Sulphid precipitation was made in a solution containing about 5 ce. concentrated hydrochloric acid for every 100 cc. of solution, which is about 2% acidity in true HCl. In the gravimetric method stannic oxid was treated with two drops of nitric acid, reignited, dried, and weighed. In the volumetric method the aluminium foil added to dispel chlorin should be added gradually in small pieces if the solution is hot, but it may all be added at once if the solution is first cooled. Excess of iodin was added in the titration and run back with thiosulphate. The end point with starch was not very sharp and inclined to be fugitive, some- what more so with the samples than with standards, leading probably to low re- sults. Both methods are rather long and tedious and are not ideal for general use. Would not consider it fitting to report results closer than thousandths of a per cent. ann 1917] TREUTHARDT: HEAVY METALS IN FOODS 51 DISCUSSION OF RESULTS. Considering the gravimetric method, the results on sample 1 show one to be correct, nine to be high, and ten to be low. Onsample 2, six results are high and eleven are low. The two lowest results on each sample have a deviation greater than —2.0. Excluding these results in obtain- ing the averages would give averages of 23.3 and 14.0, respectively. The results obtained by the volumetric are slightly lower than by the gravimetric. One correct result was reported on sample 1, and the rest were alllow. Two of the results had a deviation greater than — 2.8 and four had a deviation greater than — 2.0. On sample 2, three results were high and twenty were low. Four results had a deviation greater than — 2.3 and six had a deviation greater than — 2.0. By excluding these very low results, new averages could be obtained which would naturally be closer to the amounts present in the samples. There appears a tendency to get lower results by the volumetric method, but when the details of this procedure are observed it is quite possible to get satisfactory results. On the whole, the results on the determina- tion of tin have been highly satisfactory. Both of the above methods have been tested by the association for three years. Much work has been done on both in the Bureau of Chemistry and in commercial in- stitutions especially interested. From the experience of the referee, in connection with the work of the association and in other directions, it is felt that the association would be justified in adopting the two methods. The gravimetric determination is rather tedious, but is suitable where only occasional determinations are to be made. For the examination of a large number of samples the convenience of the volumetric method is very apparent. It should be noted that the method of acid digestion and precipitation by hydrogen sulphid is applicable to the determination of metals other than tin. Further work on tin should be in the investigation of methods applicable especially to that metal, notably, electrolytic methods and the method of Alexander, Bloomberg, and Lourie (J. Assoc. Off. Agr. Chem., 1915, 1: 259 (5) ), which was tried to a slight extent last year. Concluding the report on heavy metals, attention is called to the fact that this work requires careful, neat, and accurate manipulation. This is especially evident in the lengthy procedures for the determination of tin. This work should be undertaken by experienced analysts who are willing to devote considerable time in preliminary experimentation and who pay close attention to details. It is not easy to determine minute amounts of metals in a large excess of organic matter, and simplicity of method should not be set above accuracy. Reliance should not be placed upon close checking of duplicates, but methods should be judged by 52 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 ability to get correct results when substances of known composition are examined. Finally, attention should be called to the fact that the term “heavy metals’? does not seem to apply to a subject which will soon include aluminum, nickel, copper, and zinc. RECOMMENDATIONS. It is recommended— (1) That codperation with the associate referee on baking powder in the study of methods for the determination of lead in baking powder and baking-powder materials be continued. (2) That the gravimetric and volumetric methods for tin, tested this year, be adopted by the association as provisional. (3) That further study be made of other methods for the determination of tin. (4) That further study be made of the Gutzeit determination for arsenic tested this year, especially as to conditions affecting the evolution of arsin. If possible, a procedure should be devised which may be adopted as provisional. (5) That study be made of the various modifications of the Gutzeit method for arsenic as applied to specific substances such as gelatin, follow- ing the, procedures described in U. 8S. Bureau of Chemistry Circular 102. (6) That a study of some modification of the Marsh method for the determination of arsenic be made. (7) That the methods for the determination of copper, zinc, nickel, and aluminum in food products be made the subject of study by the associa- tion as soon as possible. (8) That the designation of this portion of the work be changed to “Metals in Foods.”’ Adjourned at 5.05 p.m. for the day. SECOND DAY. TUESDAY—MORNING SESSION. REPORT ON THE SEPARATION OF NITROGENOUS BODIES. (MILK AND CHEESE.) By A. W. Bosworts (Agricultural Experiment Station, Geneva, N. Y.), j Referee.* The work of your referee has been confined to a study of the proteins of goat’s milk and the results of this investigation will appear later as a bulletin from the New York Agricultural Experiment Station. In order that a complete study of the separation of the nitrogenous bodies in milk be made it is necessary that we fully understand the chemi- cal properties of each of the substances present. This is quite evident if we consider the possible production of nitrogenous bodies from casein and albumin by the action of the reagents used to remove them as a pre- liminary step in the study of the other nitrogenous bodies in milk. Your referee would therefore recommend that a thorough study of the chemical properties of the albumin of milk be made by the next referee. The President announced the appointment of the following committees: Committee on nominations—W. B. Ellett, Blacksburg, Va.; J. M. Bart- lett, Orono, Me.; T. C. Trescot, Washington, D. C. Committee on resolutions—R. J. Davidson, Blacksburg, Va.; C. S. Brinton, Philadelphia, Pa.; A. S. Mitchell, Washington, D. C. Committee on auditing.—C. S. Cathcart, New Brunswick, N. J.; J. K. Haywood, Washington, D. C.; W. L. Dubois, Hershey, Pa. Dr. Harvey W. Wiley and C. S. Hudson were appointed to represent the association at the second Pan-American Scientific Congress, held in Washington December 27, 1915, to January 8, 1916. 1 Presented by P. F. Trowbridge. 54 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 REPORT OF SECRETARY-TREASURER FOR THE YEAR 1914-15. By C. L. Atsspere (Bureau of Chemistry, Washington, D. C.), Secretary-Treasurer. 1914 RECEIPTS. Noy:.16:, Balancejonjhand eevee ne Pec erent eee eiran tae $161.43 Noy. 18. 1913-14 dues from 3 States (Alabama, Massachusetts, and Ver- MOD) 3 53:c)ccoemee Pod eweky Coa tie: Ee Eee Ee 6.00 Dues for the year 1914-15 from 69 Federal, State, and municipal organizations and 2 honorary members....................... 142.00 Mobalp eye eet ee sectinaatpitet clair neice mneie lee nee $309.43 1914 DISBURSEMENTS. Nov.. 18: Lelephonerealls Raleigh Motel. -c1)4 cee eee eee eee $1.20 Tips "Raletehulto tel: cascscacie stents petoatetasi br ooo eee 4.00 Dec. 7. 1,000 special request envelopes..............2-eeeeeeeeees ee ete 22.00 Dec: 19: | Postapetssedianstsaer aoe Oreo ciesmecmalocamet es eek cee 2.00 1915 Jan. 14. 6 circulars, Superintendent of Documents...................... 0.30 Feb. 2. 2 bulletins, Superintendent of Documents...................... 0.50 Heb: 8; Postage. cokers ote sete acts = ngheiseepess| sis saree er ete See eee 2.00 Beb:. 10. gPrinting 700vetreularsys.s-scncincie seer ce eect ac eee 9.75 June) i3:, sPost office boximenti and Key, shay aici sei wlatal evo itete isi taee cle eiatels 0.53 vune 19), Printing OOO Metberher dsr ey-eiait-1e ieee cies eietelotereree ene 3.75 July "25" Postfothice) box renth once eee eee popsoassan0¢ Leiden 1.00 Aug. 30. Postage, sending out announcements of meeting................ 5.00 Aug. 31. Envelopes, sending out announcements of meeting.............. 0.75 Sept. 3. Printing announcements of meeting..................2.2-e0e00e 19.75 Sept: 20:, Post office box rent. ssc cccems orioaeis shicit ona ace eC 1.00 Oct? 21. Printing s1;000 letterheads nrc nstcee hrs eeee eee eee 4.50 Nov. 2: ‘Gisheetsbristol papers. sine cite creates ets «crelcleve efeleterzia s/s seen 0.36 Nov. 4. (350 badges sect): 35 20,5) ¥. cemraeise iene inlsiai stoic bi dakyooloaly oe cee 28 .00 1,000) special) request) env.elopesin ice ciel eee cle irl iene 22.00 Bankbalances: 20.0. oo. coats e cil tele eiee Sa sere) eh « eee eee 181.04 MO Gal siege oe re He wee oOo ates visser cis tte eee $309.43 FINANCIAL STATEMENT OF THE JOURNAL OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS, RECEIPTS. 5 cash subscriptions at; $teacht. 72 2ce, fare ntss ceca a oes cate Een ee $20.00 Checks reopivedsss cn. sone eines ok dine oce enon et an, SO eee 2,131.40 Check from secretary-treasurer for Irving C. Bull’s subscription......... 4.00 Total receipts: 1.3 Pen ccealn cee oolmnak tien een ee eee $2,155.40 Interest-credited: (Sept:'80\@91b) a) alt oe oe ee ee eee 6.33 $2,161.73 1915 DISBURSEMENTS. June 3. 1,000 special request envelopes..........)....0.cceceeceeceeces $22.00 June 10. Postage: . jsccd. acme ea cet cnet ee ae eee oc 10.00 June UW. Postage, 2. joc dere se Sees eo Oe ee nee 5.00 June 12, Refund onisubseriptronses seine nee eeene nnn 1.00 June 19) Refind'on ‘subscriptions meee seine een eee ae 1.00 June 25. 1,000 special request envelopes.............00secceeeeceseeuees 22.00 June 26. 2 refunds on subscriptions at $1 each.............ceceeeeeeeees 2.00 1917] ALSBERG: JOURNAL OF A. O. A. ©. 55 REPORT OF SECRETARY-TREASURER FOR THE YEAR 1914-15—Continued. June 29. Agent’s commission on subscription............-....+++-0+000- $0.50 Secretary-treasurer, checks deposited.................--.+-+-- 3.00 Trr@, SO), Sivnmyery (E615). Seas aoseranenodGeopeposudanoooonos6oDouemcophoG 3.00 ily oe Mim ecopraphing. 2,000 eGters. «\<:.(0 7... /0« sle ela sleleleme!delmcie = lr siel-le 5.50 imiky So Siicias) (Chdn)saacebedaaseers ap boseabseMosbeoobicoo onan 6705/adc 1.00 ihniky Oo. steoreveg (GOs | oo ooops beoompeoosedsepoopsbsoaccdscor ac 1.00 diel Beh, Wao) ovyieall orikels) (ACI). 4 sspone cd sedacooneononounonaronconscd > 1.00 Jiviky: Si, Reine Onis osteeaiOmaeue seeseendees peenabdocosdccearecqaconc 1.00 Sept. 8. Williams & Wilkins Co., on account.................0.20-0+-- 1,500.00 Sete, DA » Shien) (CEM)5 64g ee os Soc pueed sees easeeoe sonsenosodosbenas’ 1 Oct. 5. 3 refunds on subscriptions at $1 each......................+-+- 3 Oni, Gy Siena G20) 554 Send cuedod ASeE ree Gnaa nee mcenceaurisooaceo acc. 0 Orel me iINGkOnISUOSCLIPLLOMN... 4 ace oe coc looocupRood lagcodccos ¢ 8.26 PE | coy sce eucll cv aveyeios ccoite ctallfehevaoe pays a seat] Wbera/steveseseussell lerepatsietceetace | tceveye ereretelous 8.30 14.91 15.04 15.04 14.99 18 47 18.39 18.40 18.35 1 Average. Mr. Davis also reported results on two samples prepared himself show- ing comparison of official method of preparing solution and modified off- cial method (no HCl). JOUR. ASSOC. OFFICIAL AGRIC. CHEMISTS, VOL. 111, NO. 1 He also used the perchlorate method in this work, 114 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 TABLE 4. Comparison of results on the use of hydrochloric acid in potash determinations with different mixtures. MIXTURE OF 1.25 GRAMS KAINIT AND 1.25 MIXTURE OF 4 GRAMS KAINIT AND 2 GRAMS GRAMS ACID PHOSPHATE ACID PHOSPHATE With HCl Without HCl With HCl Without HCl 6.69 6.54 8.09 8.22 6.47 6.68 8.07 8.26 6.66 6.64 8.10 8.29 sieleve Sisjarelebske says eroclarc al meee eee eee 8.20 sitievele: Syeyetere cle a eee 16.61 16 62 18.12 18.26 1 Average. Mr. P. L. Hibbard reported the following results on the use of HCl to the water extract. Three operators used the same mixtures, apparatus, and chemicals, but otherwise worked independently. TABLE 5. Comparison of results on the use of hydrochloric acid in potash determinations with different mixtures. NO. 1 MIXTURE.! ANALYsT 1 ANALYST 2 ANALYST 3 Mi IECOE scasadsoqe 4.67 4.59 4.62 4.77 4.60 4.68 4.57 4.60 4.73 4.72 4.60 4.70 Average?....... 4.68 4.60 4.68 Without HCl....... 4.64 4.55 4.71 . 4.62 4.59 4.72 4.69 4.56 4.69 4.66 4.60 4.72 Average?®....... 4.65 4.57 4.71 NO. 2 MIXTURE.‘ With HCl a. ener 8.77 9.08 9.12 8.85 9.07 9.00 9.12 9.07 9.25 9.10 9.07 9.06 Average®....... 8.96 9.07 9.11 Without HCl....... 9.06 9.06 8.70 9.17 8.97 8.79 9.20 9.06 9.04 9.17 9.05 9.15 Average’....... 9.15 9.04 8.92 1 Kainit, 1 part; acid phosphate, 1 part. 2 Average of all results by adding HCl to water extract, 4.65% KO. 3 Average of all results by omitting HCl to water extract, 4.64% K.O. ‘ Acid phosphate, 4 parts; K2SO,, 1 part. 6 Average of all results by adding HCl to water extract, 9.05% KO. 6 Average of all results by omitting HCl to water extract, 9.04% K2O. oe 1917] JARRELL: DETERMINATION OF POTASH 115 COMMENTS BY ANALYSTS. W. J. Jones, jr.: In commenting on the methods for the determination of potash, I would say our work here indicates that considerable additional investigation might be given to the present official method which review of the proceedings leads me to believe was adopted without as much investigation as should have been required. In using the official method we find the results decidedly influenced by the tem- perature of the water, the manner in which it is placed on the sample, the length of time the solution stands before making the determination, and a number of other factors, and so far as I can discover from the proceedings these points were not investigated before the method was finally adopted. The more we use this method the more we are inclined to feel that the difference in results found between it and the old method is due not so much to more efficient extractions and determinations of potash in the majority of samples but to multi- plication of analytical error due to marked reduction in the amount of sample taken and amount used for the determination. A number of other chemists with whom I have discussed this matter seem to have encountered difficulties similar to those we have found in this department. In many cases it is extremely difficult to get duplications on aliquots from the same solution. E. G. Proulx: The determination by the perchlorate methods (d) and (e) are on aliquots from the same solutions. The results by the modified official method on sample No. 2 average 0.16% lower than by the official method, while on sample No. 3 they are nearly the reverse, being 0.13% higher by the modified method. In two laboratory samples of fertilizers consisting of acid phosphate and muriate of potash the modified official method gave an average of 0.18% lower potash than the official method in one sample and an average of 0.03% higher in the other. The effect of the addition of 2 cc. concentrated HCl appears uncertain. It is more important to have the solution at the boiling point when the ammonia and ammonium oxalate are added. The official method using denatured alcohol gave concordant results which were slightly higher than those secured by the official method; Baker & Adamson’s analyzed petroleum ether (specific gravity 0.632), boiling point 40° to 60°C., no heavier oils or sulphur present, was used as the denaturing reagent. The denatured alcohol was very unpleasant, affecting the eyes. Results by the perchlorate methods were unsatisfactory in all samples. The final percipitates contained insoluble matter averaging 0.0017 gram in sample No. 1. In sample No. 2 the insoluble residue by perchlorate method (d) was 0.0060 gram and by perchlorate method (e) 0.0050 gram, while in sample No. 3 there was 0.0340 gram insoluble residue with method (d) and 0.0083 with method No. 2. The perchlorate method requires a greater amount of time and would certainly prove more expensive per determination than the official method. Merck’s pure perchloric acid (specific gravity 1.12), containing 0.0012 gram KCIO, in 10 cc., was used in the perchlorate method. Agart Wiberg: The perchlorate method gave in all of the three samples about 0.20% higher results than the official method, which probably was due to a trace of potassium chlorid in the potassium perchlorate with which the alcohol was satu- rated; 0.0064 gram of KClO, was found in 10 ce. perchloric acid, and corrections were made for the same. R. C. Wiley: I found in my work with the perchlorate method that 10 cc. of perchloric acid (specific gravity 1.12) was not a sufficient amount. When only 116 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 this amount was added the resulting potash percentage was often too high. I got uniformly good results by adding about 20 ce. of perchloric acid in place of 10 cc., as is called for by the perchlorate method. Of the two perchlorate methods, I prefer the Ba(OH).2 process. It seems to me that the perchlorate method is preferable to the official method for the determina- tion of potash in commercial potassium salts. I think it possible that after one became well acquainted with the perchlorate method he could make determina- tions somewhat more rapidly than by the official method. At the present price of perchloric acid, I doubt if the perchlorate method is more economical with respect to reagents used than the official method, since the platinum can be readily recovered. I do not think ordinarily the addition of HCl in the water extract influences the results. W. D. Richardson: We were unable to obtain a supply of perchloric acid and were therefore unable to complete the work on the perchlorate method. In regard to a comparison of method (a) and method (6), we do not find that the addition of 2 ec. of hydrochloric acid makes any difference in the results obtained. Our results obtained in using denatured aleohol in method (c) as compared with ethyl alcohol in method (a) are practically identical. P. L. Hibbard: The methods sent by the associate referee were closely followed. Omission of boiling the water extract with HCl in method (6) seems to be a desir- able simplification, without sacrifice of accuracy. The use of denatured alcohol seems permissible, although it has given me slightly lower results. Perhaps it may have contained a little more water, which would account for the difference. From other experiments I have made, I judge that it would be better to use stronger than 80% alcohol, whether denatured or not, for washing K2PtCle. My results with the perchlorate methods are very unsatisfactory, for reasons unknown to me. The precipitate formed from impure solutions seems to contain foreign substances not removable by 95% alcohol + KClO,4. Resolution and repre- cipitation with more HClO, gives a pure precipitate of KClO,. The Ba(OH)s process (e) for removing sulphate is simpler, but less reliable in my hands. The perchloric acid used gave no appreciable blank for K;0. Ninety-five per cent alcohol saturated with KCIO, is not a satisfactory wash for purifying the precipitate of KC1O, on account of the great variation in solubility of KCIO, with change of temperature; also, because in most cases it causes a precipi- tate of KCIO, in the filtrate which contains HCIO;. Alcohol 99% + 0.2% HClO. seems to me a much better wash. In using the 95% alcohol + KCIO, for second washing after the precipitate had once been dried and weighed, I more frequently found gain instead of loss in weight. This seems to be due to deposition of KCIO, in filter and precipitate on second wash- ing. If the wash is allowed to percolate slowly through the filter there is likely to be gain in weight, but if it is drawn rapidly through by suction, there is less proba- bility of gain. With pure KCl the perchlorate method has given me nearly theo- retical results, but when BaCl, or NaCl is present in considerable amount, results are high. On account of the necessity for removing sulphate before applying the perchlorate determination, I find this method much slower and less accurate than the platinum method; and when platinum is recovered, asit easily may be, I fancy the latter method is cheaper. : Reprecipitation was done as follows: ae 1917) JARRELL: DETERMINATION OF POTASH 117 Results of reprecipitation. {Same solution of sample No. 1 used for all these determinations.] KC1Oy rounp B Cc D gram gram gram TOSTTE pas (1h oe ee ee 0.1860 0.1850 0.1820 After reprecipitation 0.1820 0.1795 0.1810 Again reprecipitated 0.1843 0.1800 0.1822 Gooch crucibles were used to collect the precipitate. After the first determina- tion, the crucibles were washed out, dried, weighed; the filtrate evaporated with more HCIO,, precipitate filtered, dried, and weighed as at first. By this process it was expected to get rid of any insoluble or other impurities contained in the first precipitate of KClO;. Above are some of the results on one sample; not very encouraging. Numerous trials of this plan indicate that it may serve to approximate correct results when the first result is too high, due to various impurities. W. A. Davis: Hydrochloric acid does not appreciably affect the results. If any difference is caused in this way itis certainly less than that caused by other fac- tors. I think there is no doubt the difference arises from difference of sampling. With the 5 grams which I used, it is impossible to insure strict equality in the sam- ples analyzed. When accurately carried out there is no doubt that the two processes of the perchlorate method give identical results. In my experience it is more difficult, unless very special care is exercised, to obtain concordant and trustworthy results by the barium chlorid process than by the barium hydroxid process, for the follow- ing reasons: (1) It is very difficult to imsure that, after a treatment with sulphuric acid, only the quantity of barium chlorid exactly necessary for the precipitation is used. It often happens that a slight excess of barium chlorid is added, which then necessi- tates the use of considerably more perchloric acid than the 10 cc. specified in the directions in order to keep the whole of the barium in a form of soluble perchlorate. If only 10 ce. of perchloric acid is used, some of the excess of barium chlorid remains unconverted to perchlorate and is precipitated as insoluble barium chlorid when alcohol is added, so that the potassium perchlorate weighed is contaminated by barium chlorid; barium can then be detected in the potassium perchlorate after the latter has been dried and weighed. In some cases I have had the results 30% to 50% high by this cause. In such cases the impure potassium perchlorate on the Gooch should be dissolved in hot water and again evaporated with 5 cc. of per- chloric acid, the usual method of collecting the salt being proceeded with. This method generally gives approximately correct results, but frequently the result is a trifle low in such cases owing to a slight loss in the additional manipulation. In all cases, whether the barium chlorid or barium hydroxid process is used, the potassium perchlorate weighed should always be dissolved in hot water after the analysis and tested for barium by adding sulphuric acid. (2) In general I find it more difficult to get a perfectly clear solution for the final treatment with perchloric acid by the barium chlorid process than by the barium hydroxid process. After precipitating with barium chlorid it is generally necessary to allow the hot solution to stand for a few hours in order to get a pre- cipitate which filters well and gives an absolutely clear filtrate. In the barium 118 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 hydroxid process the water extract after ignition usually gives an absolutely clear filtrate at once. On this account it is more rapid than the barium chlorid process. (3) Using the barium hydroxid process, the amount of perchlorate acid pre- scribed is always sufficient to turn the whole of the barium in solution into soluble perchlorates and at the same time convert all the potash salts into perchlorate. There is never any uncertainty such as exists in the barium chlorid process, accord- ing to my experience, whether some barium chlorid may be present in the potassium perchlorate. Considerable error may be caused, sometimes amounting to 0.0050 to 0.0060 gram inthe perchlorate weighed, owing to the presence of sulphates or sulphuric acid in the perchloric acid used. If this impurity is present when the 5 cc. of perchloric acid is added to the solution used for analysis, after treatment with barium hydroxid or barium chlorid a slight percipitate or turbidity of barium sulphate appears as the solution is evaporated, owing to the interaction of the barium and sulphuric acid. I found that the perchloric acid obtainable from the dealers now frequently contains so much sulphuric acid as to give a precipitate of 5 to 6 milligrams barium sulphate per 10 ce. As a check to this source of error, the potassium perchlorate weighed in the Gooch should, after the analysis, be dissolved away by washing with about 300 ec. of boiling water. After drying at 100°C. the Gooch should be weighed and its weight compared with that before collecting the perchlorate. If there is any in- crease, the second weight of the Gooch should be used to calculate the result, the weight of true perchlorate being thus determined by the loss of weight after washing with water. In the barium hydroxid process special care should be taken not to heat too strongly during the ignition. The heat should always be well below a red heat, so as to avoid any loss of potash by volatilization. Care must, of course, be taken not to evaporate the first solution to which barium hydroxid has been added in the same hood as those in which the final solutions con- taining perchloric acid are being evaporated. The former solutions give off ammonia which would be absorbed by the perchloric acid and add to the weight of the per- chlorate finally obtained. E. E. Vanatta: Perchlorate method required more time than the official method. Four rewashings were required to secure constant weight. We have not the data to give exact cost of recovering the platinum used, but the cost, we believe, is less than the cost of perchloric acid. In regard to omitting the addition of 2 cc. of HCl and boiling, as in the official method, the results were practically identical with sample No. 3. Slightly higher results were secured with sample No. 2 when the addition of 2 cc. of HCl was omitted. However, this may have been partly due to the amount of potash washed out of the sample. DISCUSSION OF METHODS AND RESULTS. THE PERCHLORATE METHOD. The details of the barium chlorid process of the perchlorate method studied this year are the same as last year except the Davis method of collecting the final precipitate and washing the same were carried out. The barium hydroxid process was devised by W. A. Davis, Rothamsted (England) Experiment Station, who sent the details of manipulation. 1917] JARRELL: DETERMINATION OF POTASH 119 While the results by the perchlorate method show a little better agree- ment than they did last year, they are far from being satisfactory. On sample No. 1 the maximum percentage found was 12.85, while the minimum was 11.93; an extreme variation of 0.92. On sample No. 2 by the barium chlorid process: Maximum, 5.47; mini- mum 4.91; an extreme variation of 0.56. Barium hydroxid process: Maxi- mum, 5.32; minimum, 4.84; an extreme variation of 0.48. On sample No. 3 by the barium chlorid process: Maximum, 8.87; minimum, 8.48; an extreme variation of 0.39. Barium hydroxid process: Maximum, 8.81; minimum, 8.13; an extreme variation of 0.68. The above figures are taken from averages of codperating chemists. While the general average by the two processes of the perchlorate method seem to agree fairly well with the official method, it can readily be seen from Tables 1 and 2 that the results are very discordant. The perchlorate method has now been studied by the association for three successive years, and an examination of the data during these years shows that while a few chemists get good results by it, in com- parison with the official method, there are unquestionably difficulties connected with the details of the process which greatly affect its relia- bility in the hands of the average analyst. It appears that only by continued repetition can any degree of accuracy be obtained by either of the processes. This is not the case with the official method. It appears to me that this method is certainly no gain in time over the official method. In fact it takes a much longer time to carry on the process. As it cannot be applied in the presence of sulphates, they must be removed by the addition of barium chlorid or barium hydroxid. This requires an extra filtration. After this filtration, the filtrate, amounting to from 100 ce. to 150 ec., is evaporated with perchloric acid. So it takes a considerably longer time to make this second evaporation than the second evaporation of the official method. The barium hydroxid process is shorter and easier to handle than the barium chlorid process. The results by the two processes are practically the same. Since perchloric acid is expensive and platinum is easily and cheaply recovered, the perchlorate method is, in my mind, as expensive as the platinum method, especially when the longer time it takes to carry on a series of determinations by the former is considered. In justice to the method, however, I believe for the determination of potash in materials of small potash content, such as soil and plant ashes, it may be used with success. In view of the fact that the results during the past three years have been so variable, the details of manipulations so difficult to handle, that there is no gain in time over the official method by its use, and the com- ments by codperating chemists have been so generally unfavorable, it 120 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 1 would seem desirable that work on this method be held in abeyance for the present. THE USE OF DENATURED ALCOHOL FOR WASHING POTASSIUM CHLOROPLATINATE, The use of denatured alcohol, Formula 1,! for washing KePtCl. gives practically identical results in every case reported in comparison with ‘ethyl alcohol. Table 1 shows for sample No. 2 an average of 5.07% K,O by using 80% ethyl alcohol and 5.06% KO by using 80% denatured alcohol. Table 2 shows for sample No. 3 an average of 8.55% K.O with ethyl alcohol and 8.53% KO with denatured alcohol. The results reported last year using denatured alcohol were as satisfactory as results reported this year. As pointed out in the report of the associate referee last year, denatured alcohol, Formula 2,! cannot be used as a wash for K2PtCls for the reason that the pyridin, one of the denaturing agents, is precipitated by platinic chlorid. It can be tested for the presence of pyridin by adding a few drops of platinic chlorid to about 25 cc. of the alcohol. If a precipitate forms after standing about ten minutes it shows the presence of pyridin, and therefore cannot be used for washing K2PtCls. THE MODIFIED OFFICIAL METHOD. The average results by the official and the modified official methods for sample No. 2 as shown in Table 1 give the same, 5.07% KO. Sample No. 3 as shown in Table 2 gives 0.03% K.O higher by the modified official method. Tables 3, 4, and 5 show some interesting results on the use of hydrochloric acid to the water extract. They were reported by Davis and Hibbard. The figures obtained with these samples certainly show that the addition of hydrochloric acid does not lead to higher results. Table 4 shows in a case of the mixture of 1.25 grams kainit and 1.25 grams acid phosphate that the average of the three results using hydrochloric acid is practically identical with the average attained when acid is not used. In the case of 4 grams kainit and 2 grams acid phosphate, the average result with hydrochloric acid added is 0.14% lower than the result without the acid. Results reported by Hibbard (Table 5) also show practically identical results by both methods. If the addition of hydrochloric acid to the water extract in any way affects the results on these samples, the difference is so slight that it is completely obscured by the usual error of manipulation. 1U.S. Int. Rev. Reg. No. 30 (rev. Aug. 22, 1911), p. 45. 1917) AMES: SOILS 121 Therefore, since the results of last year and this year have been practi- cally the same by adding or omitting hydochloric acid, its addition is an unnecessary operation. It interferes with the volumetric determina- tion of chlorin, which in many laboratories has been made in potash solutions. RECOMMENDATIONS. It is reeommended— (1) That further work be discontinued on the perchlorate method until it has been so modified as to make it of more practical value. (2) That 80% by volume denatured alcohol, Formula 1 (U. 8. Int. Rev. Reg. No. 30 (rev. Aug. 22, 1911), p. 45) may be used for washing potassium chloroplatinate. (8) That the official method of making solutions with potash salts and mixed fertilizers be revised to read as follows: Weigh 2.5 grams upon a 12.5 cm. filter paper and wash with successive small portions of boiling water into a 250 cc. graduated flask to a volume of about 200 cc. In the ease of mixed fertilizers, add to the hot solution a slight excess of ammonium hydroxid and then sufficient ammonium oxalate to precipitate all the lime present; cool, dilute to 250 ce., mix, and pass through a dry filter. REPORT ON SOILS. By J. W. Ames (Agricultural Experiment Station, Wooster, Ohio), Referee.t The instructions for work on soils followed the recommendations made in 1914 and included a study of methods for inorganic carbon, total carbon, and lime requirement. Six samples were sent out to a number of collaborators, but results are available from five laboratories only, including that of the referee. INORGANIC CARBON. An accurate measure of the inorganic carbon content of soils is essen- tail in certain soil investigations. Since the method commonly practiced of boiling with acid does not give reliable results, it is important that a more satisfactory procedure be adopted to replace the method for carbon dioxid in soils, the removal of the method having been recommended and approved at the 1914 meeting. The object of the work on inorganic carbon was to determine the efficiency of the several methods for decomposing all the soil carbonates, whether naturally present or artificially supplied, without attacking the organic matter. 1 Presented by W. W. Skinner. 122 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 DESCRIPTION OF SAMPLES. Two soils were selected, one with a low organic-matter content and supposedly free from carbonates, while the other has approximately 3% organic matter and contains a natural supply of carbonates. To portions of each of these soils, carbonates in the form of high cal- cium limestone and dolomitic limestone were added, while one portion of each soil received no addition of carbonates. The percentage of in- organic carbon in the soil, based on the composition and amounts of carbonates added, is as follows: No. 1.—Soil supposed to be free from carbonates. No. 2.—Soil No. 1 + limestone to make 0.0290% inorganic carbon. No. 3.—Soil No. 1 + dolomite to make 0.0285% inorganic carbon. No. 4.—Soil No. 1 + dolomite to make 0.2241% inorganic carbon. No. 5.—Black clay soil containing natural supply of carbonates. No. 6.—Soil No. 5 + dolomite to make 0.1425% inorganic carbon in excess of that present in soil No. 5. The limestone and dolomite added to the soils was ground to pass a 100-mesh sieve and the carbonate content determined. The soils were reduced in a porcelain ball mill to a fineness of less than one-half milli- meter. After mixing the finely ground limestone with the soil, the mix- tures were reground for two hours. The analytical results for the soils to which carbonates were added indicate that a thorough mixture of the limestone materials and soil was secured by this treatment. In the prep- aration of soil samples for analysis, carbonates naturally present would not be more finely reduced than the carbonates in the several mixtures of soil and carbonates. INSTRUCTIONS FOR INORGANIC CARBON. Use 20 grams of samples 1, 2, 3, and 5; 10 grams of samples 4 and 6. Method A.—Boiling soil with 1 to 50 hydrochloric acid under reduced pressure as suggested by Marr. Place soil in suitable flask which will withstand vacuum of approximately 70 cm. Add 80 ec. carbon-dioxid-free distilled water, mix thoroughly, connect to apparatus and start suction. When air has been removed from apparatus and vacuum of 65 to 75 em. obtained, run in through separatory funnel 20 cc. dilute hydrochloric acid (2 ec. hydrochloric acid (specific gravity 1.19) to 18 ce. of water). Boil for 30 minutes. Bottom of flask should be about three-fourths of an inch above gauze protecting it from free flame. If liquid is drawn up condenser tube, flame should be lowered. Carbon dioxid evolved is absorbed in a sodium hydroxid solution made from sodium, 25 cc. of 4% sodium hydroxid and sufficient water to cover glass beads in absorbing tower. Relieve vacuum, wash out contents of absorbing tower with 250 ce. carbon-dioxid-free water, using 25 ce. portions, and titrate. \ 1J. Agr. Sci., 3: (2) 155. 1917) AMES: SOILS 123 Titration: For the assistance of those who have had no experience with the Brown & Escombe double titration method, the following details are given: Add 1 ce. phenolphthalein to solution and run in normal hydrochloric acid until pink color begins to fade, then add N/20 hydrochloric acid to complete disappear- ance of color. Take no account of N/1 or N/20 acid used. When end point is reached, add two drops of methyl orange solution (1 gram per 1,000 cc.) and titrate with N/20 hydrochloric acid until lemon color of alkaline methyl orange just ap- proaches distinct pink color. Take reading of N/20 acid and subtract correction obtained from blank determination run under same conditions. 1 ec. N/20 hydro- chloric acid = 0.0022 gram carbon dioxid. In this titration it will be necessary for each analyst to establish and adhere strictly to a constant end point for both indicators. It will be well for those not familiar with the titration, to practice on a 4% sodium-hydroxid solution containing small amount of sodium carbonate. Method B.—Modified procedure proposed by McIntire and Willis! using 1 to 10 hydrochloric acid and constant agitation at room temperature. Method C.—Boiling soil with 1 to 10 hydrochloric acid for 30 minutes. Carbon dioxid evolved is absorbed and titrated as in Methods’A and B. BLANK DETERMINATION ON SOIL AFTER REMOVAL OF CARBONATES, In order to apply correction for possible action of acid on organic matter in three procedures used, it will be necessary to remove carbonates from sample 5. The blank determination on sample 5 after removal of carbonates will also apply to sample 6. Sample 5 originally contained carbonates. Sample 6 is same soil as 5, but has had carbonates added to it. Determination of inorganic carbon in sample 1, which, so far as is known, does not contain carbonates, will serve as blank for samples 2, 3, and 4. Removal of carbonates from sample 5.—It is suggested that a method similar to the following be used. Place 20 grams of soil in a 250 cc. beaker, add 100 ce. of 1 to 10 hydrochloric acid and stir. Allow to stand for two hours, or longer if necessary, stirring occasionally. Then filter through blue ribbon filter paper folded in Biichner funnel. The paper can be folded into a cylindrical shape to fit the funnel by forming it over cork of the same size as inside diameter of funnel. Wash soil a few times to remove as much acid as possible, using carbon-dioxid-free water. Continue suction until all excess moisture has been drawn off soil, then transfer soil from paper to apparatus and . proceed to evolve carbon dioxid as outlined under the three procédures. RESULTS FOR INORGANIC CARBON. The data for inorganic carbon are presented in the following tables: Table 1 contains results for the association samples. Results for these same soils by modifications of methods outlined in instructions are given in Table 2. ; 1 J. Ind. Eng. Chem., (1915), 7: (3) 226. 124 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 TABLE 1. Inorganic carbon, per cent in air-dry soil. SAMPLE 1 SAMPLE 2 ee Method! Method! A B Cc A B Cc C. J. Schollenberger, Ohio.....| 0.0006 | 0.0006 | 0.0096 | 0.0303 | 0.0282 | 0.0372 0.0009 | 0.0006 | 0.0096 | 0.0294 | 0.0288 | 0.0372 OF000683| Pe eemee: 0.0096 | 0.0288 | 0.0282 }........ oceanic oe on 0200935) 0203034). er cere clae eee Average.................| 0.0007 | 0.0006 | 0.0096 | 0.0298 | 0.0284 | 0.0372 0.0007 | 0.0006 | 0.0096 | 0.0007 | 0.0006 | 0.0096 Less\iblankeMichig ante err a eerris eerie eine (MUP TESBoacanea|leqnccos: Siianitrs| aShheed |paaubeae OL0215 | .....ceulece ene Average? : ii clef cic kictere|| Cli yorae al belie ai eteie[ectalete stare O:0246 lls. fscteiere| Ceerenerstee 1A, Marr method; B, McIntire method; C, boiling with 1 to 10 hydrochloric acid for 30 minutes. 2 Blank on soil not stated. 1917) AMES: SOILS 125 TABLE 1—Continued. SAMPLE 3 SAMPLE 4 ANALYST Method! Method! A B Cc A B Cc C. J. Schollenberger, Ohio.....| 0.0294 | 0.0267 | 0.0372 | 0.2172 | 0.2142 | 0.2225 0.0300 |} 0.0258 | 0.0372 | 0.2172 | 0.2124 | 0.2232 0.0300) |0.0270) |\7..5.... 022208) OR21245 Pee i. OFO2Z88 7100276) |e 022220) |, 022130" ean «1 ee rsra iets 'al| rare te taveraval| ABs jo seis See ON 22323 | Re crenreee|| creteivers rs Average...............--] 0.0296 | 0.0268 | 0.0372 | 0.2201 | 0.2130 | 0.2229 0.0007 | 0.0006 | 0.0096 | 0.0007 | 0.0006 | 0.0096 Less blank.......... pone 0.0289 | 0.0262 | 0.0276 | 0.2194 | 0.2124 | 0.2133 L. G. Willis, Tennessee........ 0.0330 | 0.0275 | 0.0410 | 0.2115 | 0.2132 | 0.2253 0.0316 | 0.0289 | 0.0410 | 0.2107 | 0.2121 | 0.2287 Bah pear. QXO292 7 |e sere te oe Orel O5nlserersertere Reece (O)C07( 0) ee ae See roe el oencatad| seo doaoo Average.................| 0.0323 | 0.0281 | 0.0410 | 0.2111 | 0.2119 | 0.2270 0.0060 | 0.0022 | 0.0146 | 0.0060 | 0.0022 | 0.0146 sess) blank: vac ss02 004 - 0.0263 | 0.0259 | 0.0264 | 0.2051 | 0.2097 | 0.2124 W. H. Sacks, Illinois..........] 0.0278 | 0.0172 | 0.0370 | 0.2070 | 0.1621 | 0.2163 (Reported by E. Van Alstine) | 0.0278 | 0.0185 | 0.0383 | 0.2082 | 0.1537 | 0.2226 by aeyebecess ys OFOIS Tale sects leone ce | Onl 531s |FO2208 Average.................| 0.0278 | 0.0181 | 0.0376 | 0.2076 | 0.1563 | 0.2199 0.0008 | 0.0024 | 0.0101 | 0.0008 | 0.0024 | 0.0101 Messiblanke 0.2811 0.2724 | 0.2224 | 0.2775 0.0014 | 0.0023 | 0.0079 AV OYA GO Nace cs ss csteesie ss egsiblankteeeere ets ieeiete 0.2710 | 0.2201 | 0.2696 W. L. Latshaw, Kansas.......) 0.1285 |........ 0:2768 ||. .-aeer 0.3408 “achedeiet 0.2768 |........| 0.3392 sig staisiatars 0.2768 |........] 0.3380 scdoe Gnu beoahesalpavacoor 0.2768 | teeth once INE eas cooso dn ecoooo|) Uses loo gcc cos O5276831 5 een 0.3390 SOS Rone se Gorncor | WAbs: Oil me ouncoglresia ace 0.0346 Tess! blanke creep ieecictetele «| O slceooil| tcrsteteretel 0.2768) || f sce 0.3044 ©: BF. Jensen}, Michiganie....-| OLl2085) ule ces clkeiecetie = 0.24810) |i 582 eect SEN Ross tes 3 0.2185) ic. on: | eee AV OLRZO4 sc). alsin = 1 0.1303 |.. 0.2833). sents] Coser 1A, Marr method; B, McIntire method; C, boiling with 1 to 10 hydrochloric acid for 30 minutes. ? Blank on soil not stated. 1917] AMES: SOILS 127 TABLE 1—Continued. BLANK ON SAMPLE 5 AFTER EXTRACTION WITH 1 To 10 HYDROCHLORIC ACID ia Method! A B Cc reschollenberrern Ohi overnc| seis /-/arstere ote sil erste «ieteteleleleyanniotstede 0.0096 Nera ok eopn kh Sasi cteleerotaw 4 raterete 0.0102 I NOENR a dacdoows 5 cos bed lbp boo opeee el oaDa speBaoscoesEsods 0.0099 L. G. Willis, Tennessee........ 0.0023 0.0009 0.0179 0.0020 0.0009 0.0168 PNVCTAR Ca creleiscis cefersi ote eres 0.0022 0.0009 0.0173 W. H. Sacks, Illinois.. 0.0007 0.0020 0.0077 (Reported by E. Van Alstine) 0.0021 0.0025 0.0081 FAV CEAG Cr erieoetslstke sees = 0.0014 0.0023 0.0079 1A, Marr method; B, McIntire method; C, boiling with 1 to 10 hydrochloric acid for 30 minutes. TABLE 2. Miscellaneous results for inorganic carbon by modifications of Methods A, B, and C. ANALYST SAMPLE| SAMPLE| 8AMPLE| SAMPLE) SAMPLE |SAMPLE 1 2 4 6 L. G. Willis, Tennessee: Method A, using 3 to 100 hydrochloric acid. Average Less blank on soil C. J. Schollenberger, Ohio: Method B, shaking 1} hours. ......... Average Method C, carbon dioxid measured... . Average Less blank on soil W. H. Sacks, Illinois (reported by E. Van Alstine) ; Boiling 2 minutes with 1 to 1 hydrochloric acid; carbon dioxid meas- ured. Average Less blank on soil 3 Hae 5 0||sb0000|(aea006 0.2182)......|0.2797 agepetorecel aise terekallevetsiszers 0.2162)......|0.2840 sap d0)|lucoocu|\aesoae 0.2172)......|0.2819 anced \sonsae\soceds 0.0060)......|0.0022 Sodnod||esaqas'lacsdon 0.2112)......|0.2797 e155 | |Gn 25 5s|lbosgad sence Nelo 0.2748 §90009||soa8aallsansaallooapsalloosous 0.2688 Footon|lodasss| paocdclledaads annals 0.2718 0.0103/0 .0369|0.0373/0 .2313|0. 1608/0. 2980 ANU El oe cocllpadebeleaasod|leconcellscooor WA US) ona potane Roose olloooocd lansaes 0 .0093/0 .0093)0 .0093/0 .0093]0 .0100/0.0100 0 .0276)0 0280/0 .2220/0. 1508/0. 2880 0.02880 0316/0. 2203/0. 1482/0 .2804 0 .0288)0 .0311/0.2203)/0. 1482/0. 2823 0.0036, 0 .0288/0 .0314/0 2203/0. 1477/0 .2814 0.0041/0 0041/0 .0041/0.0045/0.0045 0.0041 128 ASSOCIATION OF OFFICIAL AGRICULTURAL cHEMIsTS [Vol. III, No. 1 COMMENTS OF COLLABORATOR. E. Van Alstine: Comparison of figures would indicate that the Marr method gives results perhaps more nearly correct, for with sample 1, which is supposed to contain no carbonates, it gives much the lowest results, indicating that it decom- poses the least amount of organic matter, while with soils known to contain car- bonates, it gives results that compare closely with the method of boiling at atmos- pheric pressure with hydrochloric acid. The McIntire method, on the other hand, not only seems to decompose a fairly large amount of organic matter, but with soils containing large amounts of car- bonates the results are apparently too low, indicating that the method not only attacks organic matter but does not decompose all of the carbonates present. Boiling for 30 minutes with 1 to 10 hydrochloric acid at atmospheric pressure decomposes more organic matter than it should; but when this, as found in samples 1 and 5 after removal of carbonates, is applied as a correction, results on the other samples compare closely with all except the MacIntire method. The same is true with the method of boiling for 2 minutes in 1 to 2 hydrochloric acid, but much less organic matter is attacked when the same soil is boiled for 30 minutes in a weaker solution. A method, to be practicable, of course must be one with which one need not apply a correction other than that for impurities in the reagents, since it is evident that one can not correctly apply asa correction the inorganic carbon indi- cated to be present in a soil known to contain none, to other soil containing lime- stone and which may also contain organic matter more easily or less easily attacked by the reagent. It is also very evident that in practice one can not remove car- bonates from a soil ‘supposedly’? by any method, then treat the residue with re- agents to be used in the organic carbon determination, applying the apparent car- bonates found in this way as a correction. If this is done, the question at once arises, What method shall we use to free the soil of carbonates and at the same time leave the organic matter unattacked, to be acted upon by the reagents in the regular inorganic carbon determination? If we can use one of these methods for inorganic carbon without applying a cor- rection other than that for impurities in the reagent used, then the Marr method has an advantage, since it attacks organic matter much less than any other method does. One thing which should be tested thoroughly is whether or not the Marr method will liberate all the carbon dioxid from dolomitic limestone which is found in many soils in the northern part of this State, and undoubtedly in other States as well. I may say, in this connection, that the method of boiling for 2 minutes in 1 to 2 hydrochloric is much quicker, more easy to manipulate, and for comparative results for most soils is accurate enough so that one seems justified in using it in preference to the Marr method. In certain peat soils which contain a certain amount of acid decomposed organic matter, I have found that results by the Marr method are more trustworthy. L. G. Willis: I believe, judging from the color of the acid extract of sample 5 and the slight activity on the organic matter due to treatment in Methods A and B, that the use of 1 to 10 hydrochloric for extraction causes some decomposition of the organic matter, and the blanks obtained are, therefore, too low. I question whether the variations due to this probable error in the blank are constant for all methods. C. J. Schollenberger: The results this year confirm those of 1914, in that the MacIntire method for inorganic carbon does not give satisfactory results with soils containing any considerable quantity of natural carbonates. The cause is probably the practical impossibility of grinding the sample to a sufficient degree of fineness 1917] AMES: SOILS 129 to be completely decomposed by cold acid, or the well-known resistant character of dolomitic limestone when not ground to an impalpable powder. The Marr method has given fairly satisfactory results; very good, indeed, when compared with Method C, using the double titration procedure for the estimation of the evolved carbon dioxid, with the single exception of sample 6. The somewhat low results by the Marr method on this sample indicate that it may sometimes be advisable to increase slightly the strength of the acid over the standard 1 to 50. When the results by Method C, using the double titration method, are compared with those for Method C, measuring the carbon dioxid, the results are found to be lowin the cases of samples 4, 5, and 6. This is no doubt due to the well-known inaccuracy of the double titration method when used for the estimation of rather large quantities of carbon dioxid. Repeated trials with both Methods, A and C, using about 0.2 gram Iceland spar and estimating the evolved carbon dioxid by the double-titration method, gave a recovery ranging from 95% to 99%. Ina trial with the same sample of Iceland spar, using the Marr apparatus, 1 to 50 hydrochloric acid and boiling under vacuum of 65 em., but dispensing with the bead tower and substituting a Meyer bulb tube with barium hydroxid solution, filtering, and titrating the precipitated barium carbonate, a perfect recovery was obtained. Lack of time prevented a further study of this procedure as applied to soils. W. L. Latshaw: In Method A, boiling under reduced pressure, we found that half-pint milk bottles were very satisfactory to withstand the pressure. This method of determining carbonates in soil is very satisfactory, and we plan to use it in our survey work. DISCUSSION OF RESULTS. The results reported by collaborators who had previous experience with the methods from special studies on determination of soil carbonates having been made in their laboratories show a fairly close agreement where the same method was used, after the blank determination for each method is applied. Soil 1 is from cultivated plots of the Ohio Experiment Station and has never been limed. This soil is derived from sandstone and shales, and, so far as is known, does not contain either a natural or artificial supply of calcium or magnesium carbonates. Boiling this soil for 30 minutes with 1 to 10 hydrochloric acid gives a higher figure for inorganic carbon than is shown by boiling with 1 to 50 acid under reduced pressure (Method A), or with 1 to 10 hydrochloric acid at room temperature (Method B). This points out the effect of boiling with acid on the organic matter of the soil in the methods ordinarily employed for the determination of inorganic carbon. Boiling soil 1 with 1 to 1 hydrochloric acid for 2 gives a lower figure than is obtained by boiling with acid one-tenth as strong for 30 minutes. This indicates that the action on the organic matter is due more to the heating at 100°C. than it is to the strength of the acid. The blank obtained by boiling soil 1 for 2 minutes 1J. R. Cain. J. Ind. Eng. Chem., 6: 465. JOUR. ASSOC. OFFICIAL AGRIC. CHEMISTS, VOL. III, NO. 1 130 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 with 1 to 1 hydrochloric acid is much greater than the blank shown by boiling with 1 to 50 acid with reduced pressure (Method A). Results for soils 2, 3, and 4, in which the added carbonates were in a finely divided condition, show a more complete decomposition and recoy- ery of carbonates by both Methods A and B than is obtained from soil 5, which contains a natural supply of carbonates. In the ease of soil 5, having a natural supply of carbonates, and soil 6, which contains added carbonates from dolomite in addition to the natural content of soil 5, the results obtained by boiling with 1 to 50 — acid under reduced pressure agree with those by boiling for 30 minutes with 1 to 10 acid after subtracting the blank, and are considerably higher than results obtained by treating with 1 to 10 acid at room temperature (MacIntire method), which has not been sufficient to decompose the calcium and magnesium carbonates. The blank determination on the soil previously extracted with acid to remove carbonates gives some indication of the action of the procedure employed on the organic matter, but there is no certainty that by the treatment with acid for the removal of the carbonates a portion of the organic matter which would be easily acted upon is not removed by the acid treatment. That there is a considerable amount of carbon from organic sources obtained by boiling the soil with 1 to 10 acid for 30 min- utes after removal of carbonates is evident from the large blank obtained. When this blank is applied as a correction to results by this method, the figures obtained agree closely in most instances with the results by Method A. It is not probable that there is any appreciable action on the organic matter by either Methods A or B. The high figure obtained for the blank in some instances can no doubt be considered partly as a blank on the manipulation of the process rather than being entirely due to the activity of the acid on organic matter. It will be impossible to decompose the carbonates in soils without producing some slight action on the organic matter. This will be greater for some soils than others, depending upon the nature of the organic matter present. Making a blank determination and applying this as a correction is impracticable. The most satisfactory procedure will be to employ a method which will decompose all the carbonates and at the same time have the least activity on organic matter present. The Marr method for decomposing soil carbonates seems to fulfill this condition better than any method which has thus far been proposed. The results as a whole indicate that the Marr method is more efficient than the McIntire method for decomposing calcium and magnesium carbonates, whether naturally present or artificially supplied. The estimation of carbon dioxid evolved from carbonates by titration is not altogether satisfactory. In the double titration procedure there 1917] AMES: SOILS 131 are sources of error which tend to give low results, especially when the carbonate content of the soil is fairly large. Results by C. J. Schollen- berger show a more complete recovery of inorganic carbon from soils 4, 5, and 6 is secured by decomposing the carbonate after absorption in sodium hydroxid and measuring the carbon dioxid. Instead of absorbing carbon dioxid in sodium-hydroxid solution and making a double titration, more accurate results will be obtained by absorbing carbon dioxid in barium hydroxid, using a Victor-Meyer ab- sorption tube and determining the carbonate by standard methods, either titrating the barium carbonate after filtering and washing, or by making a gravimetric determination. TOTAL CARBON. INSTRUCTIONS TO COLLABORATORS. For determination of total carbon use samples 1 and 5. ' Transfer 3 grams of soil into a short-necked Kjeldahl or other suitable flask and connect to same apparatus used for determination of inerganic carbon. Run into flask through separatory funnel 10 cc. chromic-acid solution containing 3.3 grams chromic acid, then 50 ec. concentrated sulphuric acid. The mixture is boiled for TABLE 3. Total carbon results. sor 1 sort 5 Combustion with sulphuric and chromic acid; carbon dioxid titrated: C. J. Schollenberger, Ohio............... 0.8000 2.7280 0.8000 2.7360 0.8020 2.6800 0.8040 2.6960 Bee Ee ct AAI 2.7460 Ty ga ee cociag DERE CHG SECIS 0.8015 2.7172 Weapls: Wallis; @ennessee! taaepicos > = otc - 0.8380 2.7270 0.8300 2.7390 0.8350 2.7370 0.8490 2.7370 0.8430 2.7310 OFS270FO whist Le eee IBV CTR ECs oes ccaieiefctesierni wie eerste jsrenieietelate 0.8370 2.7340 Combustion with sulphuric and chromic acid; carbon dioxid measured: C. J. Schollenberger...................-. 0.9010 2.9030 0.8860 3.0050 hovsera/pisteioagoe ere reierers 2.8940 AVICL RE Cue ita Sen sions oie bine pietoslic 0.8940 2.9340 Combustion with copper oxid in furnace, carbon dioxid weighed!................... 0.8800 2.9525 0.8800 2.9525 1 The products of combustion were passed over heated copper oxid in a second tube; the residues we tested for carbonate in all cases, with negative results. 132 ASSOCIATION OF OFFICIAL AGRICULTURAL cHEMISTS [Vol. III, No. 1 30 minutes, during which time a moderate current of carbon-dioxid-free air is drawn through the boiling mixture. Carbon dioxid may be absorbed and titrated as directed under instructions for inorganic carbon. About 5 drops of a 10% solution of sodium thiosulphate should be added to alkali before titration to overcome any effect caused by carrying over of chromic-acid compounds. It will be well for each collaborator to compare method with combustion in furnace or other method for inorganic carbon used in his laboratory. COMMENTS OF COLLABORATORS. C. J. Schollenberger: As was the case in 1914, the results for total carbon by the method of wet combustion with sulphuric and chromic acids, estimating the evolved carbon dioxid by double titration, are considerably lower than those obtained by a furnace combustion, absorbing and weighing the evolved carbon dioxid. When a gasometric method is substituted for the double titration procedure, the results obtained by the wet combustion compare fairly well with those obtained by furnace combustion and weighing. Upon comparison of the results for inorganie carbon by Method C, using the double titration and the gasometric methods for estimation of the evolved carbon dioxid and for present purposes considering the latter to be correct, it is found that the average percentage of recovery was about 97.7. A similar comparison of the results for total carbon indicates a percentage of recovery of only 91.8. This would indicate either incomplete absorption of the evolved car- bon dioxid, or possibly interference due to the presence of large amounts of sulphate; it was noticed that one-third to one-half the alkali in the absorption tower was invariably neutralized by the acid fumes from the boiling acid in the flask. DISCUSSION. While the results for total carbon are insufficient to permit of any definite conclusions being drawn from the work done so far, it is evident that combustion with mixture of chromic and sulphuric acids and deter- mining the carbon dioxid by titrating the absorbing alkali solution gives lower results than are obtained by measuring carbon dioxid resulting from the combustion. The amount found by measuring the carbon dioxid evolved from the wet combustion agrees closely with that found by combustion of soil in furnace and weighing. This indicates that boiling with a mixture of chromic acid and sulphuric acid, as outlined in the instructions, is sufficient to oxidize the organic matter, but that the lower results are due to errors in the double-titration procedure, when used in combination with the wet-combustion method. The work on organic carbon was taken up following the recommenda- tion made by the referee on soils for 1913. It does not seem advisable to continue the work, since the standard method of determining organic carbon by combustion, either in current of oxygen or by mixing with copper oxid, and the optional official method are efficient. Since the figure for total carbon obtained by whatever procedure is employed must be corrected for carbon from carbonates which may be present in the soil, it is important that this correction be based on a more accurate determination than is commonly practiced. 1917] AMES: SOILS 133 LIME REQUIREMENT. The instructions sent out suggested that those interested compare two of the more recent of the numerous methods which have been proposed for the estimation of lime requirement, with the Veitch, Hopkins, or other method on soils of known history available for the purpose. It was also stated that sample 1 of the soils for work on inorganic carbon be used. The two methods referred to are the Hutchinson-MacLennan! method and a modification of that published by McIntire.” In both these methods the calcium absorbed from the solution of cal- cium bicarbonate is taken as a measure of the lime requirement. Some of the collaborators made lime requirement determinations on the association samples to which calcium and magnesium carbonates had been added as stated in Table 1. The results obtained in study of lime requirement methods are given in the following tables: TABLE 4. Lime requirement results. {Average per cent calcium carbonate required.] METHOD ra! SOIL ao a> = | vse Hutchi Modified = utchinson- . Mo 4 MacLennan| McIntire McIntire! Jones grams W. H. Sacks, Illinois (re- ported by E. Van Alstine). Oo meee or oO 0.0735 020033) {2 eeeeene 20 0.0200 to to: 2 |e se 0.2735 0:0266"" |e es: G. L. Willis, Tennessee......}] 1 ]|...... 0.3565 TEEOSO NaN icon. cdee Al ance W. L. Latshaw, Kansas..... 1 ae ONTTOOM | Fee eee ch |S cs fetes: ol sate-c ncteele 7 rao GLOSS" |i Secesseceleces Fisieceval| e sresisie’s sreye Bialivecitee QROGOO Me fee meres ee ae cvcters tesa eo relencteyeyeters CS eee QROS2O i ers [A AES ranshacsi| a eteietereretele 3 igaeriog SUC AN nc ceed Gene Meee onc ao belo Gules ts 200250) | |More on | eras ccereieiotacall frops state cnarare C. J. Schollenberger, Ohio...| 1 | 20 QuZTOO MEE Fe eo 43.4. eel en heey 1 Tt} SaccksnoG.o > ORS TOO ie cites eal Miser eects OyaptreJerisen, (Michigan. saoe | ella pre mayors ie oe Orel RN tetetevat sal seavereeic vere ore 0.2007 yal petal naisolke aiteal o ceenetrael Geico 0.0963 Bilateral er aero ees | Pete ban Shere 0.1766 0.1445 4 lee ear ees teeee eel ten cteeten eer] coeects ene iste to 0.1766 Sn nob be aaa scan | SONS r ar aa) meAaa Ss eer, 0.0482 Gill ites So lhe esac | eile sash Sal] erage ecacoelee 0.0321 _ 1 Excess calcium carbonate determined by boiling for 2 minutes with 1 to 1 hydrochloric acid and meas- uring carbon dioxid evolved. , 2 Per cent calcium carbonate given up by soil. 1 Chem. News, Aug. 7, 1914, vol. 110, No. 2854. 2 J. Ind. Eng. Chem., 7: (10) 864. 134 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 TABLE 5. Lime requirement results on miscellaneous Illinois soils by methods indicated. (W.H. Sacks and E. Van Alstine, analysts.) [Average percentage calcium carbonate required.] METHOD SAMPLE NO. Bor vusEepD | Hutchin- : Modified | Hopki poe McIntire pr cveriterd ban essai ie | a na. le. on 1745 (0 to 7 inches).......... 20 | 0.148 | 0.083 + to 0.384 0.1905 | 0.0185 1746) (7 tov Gianches) sa-csm ects ee |) Os S41 ba tee teenies celine recieves 0.2905 TOM i cvevers oiezs 0.240 + to 0.450+ |........]........ 1747 (18:to40'inehes)*.25 52 cl) (20) | ORLOGE GIs Sc sejoteterc cress aystaioks ote: s\oif/= cpesteleverel Cesena 10 | 0.795 | 0.303 + to 0.585+ | 0.7034 | 0.8090 5 | 0.845 | 0.415 + to 0.560+ | 0.8432 |........ 1748 (0 to 7 inches).......... 20) | O90) i iocse coyote smecrtesoelevece Ree 0.0250 10 | 0.200 02322 ysis varevctereval| eretoveretetee 1749 (7 to 18 inches)......... 20 MeO e elo als acters terete rete leis) | etetotemetcte 0.123 1750 (20 to 40 inches)........ 20) 1) (0 S600 hil. .0 atecyetetetsieeese sierra Sere ees 0.2815 1751 (O)toiinches) sce oe.cctee = |e v2 a AO ML2GOiN| ree ete cts Sots alsieret-rereneie | ereee eit 0.005 1 Soil and 100 cc. calcium bicarbonate solution evaporated to a thin paste and excess carbonate deter- mined by boiling for 2 minutes with 1 to 1 hydrochloric acid and measuring carbon dioxid evolved. TABLE 6. Lime requirements obtained by varying conditions of McIntire method. CALCIUM ae rmEATMENT ae POUNDS OF SOIL grams pounds 10 Evaporated to thin paste immediately.............. 7,500 10 Evaporated to dryness immediately................ 8,200 20 Evaporated to thin paste immediately.............. 5,750 20 Evaporated to dryness immediately. . ; 6,300 10 Evaporated to thin paste immedi ately; to ‘dry ness 12vhours latent... aeet ex «seus ce ois k ee aah eee 8,900 10 Soil and solution in contact 12 hours before evapo- rating to dryness COMMENTS OF COLLABORATORS E. Van Alstine: The Hutchinson-MacLennan method gives much higher results on soils with a small lime requirement, in this case surface soils, than does the Hop- kins method; the two methods giving more nearly the same results on subsoils with a large lime requirement. This is also true of the McIntire method. By varying the amount of soil used, the Hutchinson-MacLennan method shows in every test on three soils, 1745, 1747, 1748, that with smaller amounts of soil the lime requirement appears to be higher, the greater variation appearing on soils with a high lime re- quirement. ‘Thus on soil 1745, varying the size of sample from 20 to 10 grams, causes 1917] AMES: SOILS 135 -a variation from 0.148 to 0.170; on sample 1747 varying the size of sample from 20 to 5 grams causes a variation in apparent per cent of calcium carbonate required from 0.706 to 0.845. Sample 1 sent out this year for referee work shows this same sort of variation. When 20, 10, and 5 gram samples were used, the per cent of cal- cium carbonate required varied from 0.156 to 0.290. The MacIntire method shows the same sort of variation, the per cent of apparent calcium carbonate requirement of sample 1747 varying from 0.703 to 0.843 with 20 and with 10 gram samples. It would seem from this fact that when a soil is treated with calcium bicarbonate solution or with calcium carbonate in water or in a salt solution, as I have found by previous work along this line, the amount of carbonate which the soil will decompose and the amount of calcium which the soil will extract and hold is not constant, but varies with the ratio between the amount of soil and amount of carbonates pres- ent. This would mean that a soil with a high lime requirement would decompose a larger amount of carbonate, thus in effect weakening the carbonate solution, than would a soil with a low lime requirement, so that in order that the results on such soils be strictly comparable, one should use a larger amount of carbonate with a soil high in lime requirement than with one which has a low requirement, which of course would not be practicable. Moreover, there is no means of knowing the prop- ‘er amounts of soil and solution to be used in order that the results may represent the truth in regard to the calcium carbonate any soil requires. It is evident that if either method is to be used, arbitrary amounts of soil and solution must be adopted; this then would give only comparative rather than absolute results, and fall short of our needs for an exact method as do other methods now in use. As it is hoped that the method represents what actually takes place in agricultural practice, then it would seem that one should use with the soil sample only as much calcium car- bonate as would represent the amount most often used for acid soils. If one takes 2.000,000 pounds as the standard weight of the surface soil, then to represent a ton application of limestone to the acre would require but 10 milligrams of calcium carbonate, or 10 cc. of the N/50 bicarbonate solution, for a 10-gram sample; for perhaps the most common application of limestone, 2 tons to the acre, this would be but 20 milligrams calcium carbonate or 20 cc. of bicarbonate solution for a 10-gram sample. This would clearly give very much lower results than are obtained by the amounts used in the method as outlined (250 cc.) which represents 25 tons to the acre with a 10-gram sample. The Hutchinson-MacLennan method shows a lime requirement on soils which have a considerable amount of calcium carbonate in them. Sample 2, wnich none of the methods used for inorganic carbon shows to contain less than 0.14% calcium carbonate, has, according to this method, a caleium-car- bonate requirement of 0.095%, equivalent to an application of nearly a ton of lime- stone to the acre; while sample 5, which every method for inorganic carbon shows to contain 0.85% or more of calcium carbonate, is shown to have a calcium carbonate requirement of 0.02%, or 400 pounds per acre. This is in accordance with what is indicated by varying the size of the sample, that the greater the amount of calcium carbonate the soil is treated with the more it will decompose and the more calcium it will absorb. MclIntire’s method shows this same thing, although there is a greater variation in the results obtained. With this method it seems clear that one must subtract the calcium carbonate remaining after the reaction from the sum of the calcium carbonate added in the bicarbonate solution and that indicated to be present in the soil by the method used for determining calcium carbonate in the residue after the reaction, even though one is working with an acid soil, since as much carbon dioxid would be driven off from the soil alone after treatment with bicarbonate solution as would be before treatment. 136 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Making estimations in this way, the McIntire method shows soil 2 to have a calcium carbonate requirement of 0.291% (2.9 tons per acre), while results on soil 5 indicate a lime requirement varying from 0.07% to 0.27%. I may say, in this connection, that results obtained on lime requirement from the calcium carbonate in the residue, also determination by the McIntire method of aération were not satisfactory, as is shown by comparing results obtained in this way with those in the column under modified MeIntire method. Results are much closer when the remaining carbonate is determined by boiling with hydrochloric acid and measuring the carbon dioxid evolved. Because of the fact that the double titration for bicarbonates is not altogether satisfactory, especi- ally in the hands of chemists not thoroughly familiar with it, the end points not always being distinct, and because with the double titration there are two chances for error, we favor the method of measuring the carbon dioxid which should give results as accurate as those obtained by accurate titrations. The McIntire method does have an advantage over the Hutchinson-MacLennan in one respect, however. Thus, results for sample 5 show that the soil has actually used up some of the calcium carbonate, but after the reaction there is left in the soil more calcium carbonate than was added, so that while the treatment caused the soil to absorb calcium and decompose the carbonate, there is in the soil from the begin- ning more than this amount of carbonate and the soil can not be said to be in need of an application of limestone. Sample 2, which apparently from the inorganic carbon tests does not need an application of limestone, does have, according to the McIntire method, a large lime requirement, and this requirement is more than great enough to require all the limestone in the soil, as found by the McIntire test for inorganic carbon. In this one respect, that with soils comparatively high in carbonate content there is opportunity to see whether or not the carbonates in that soil are sufficient to satisfy its apparent lime requirement, the method proposed by McIntire seems to be preferable to that of Hutchinson and MacLennan. Yet, even with this ad- vantage, the lime requirement which it indicates must depend upon the use of arbitrary amounts of soil and bicarbonate solution in the test, a condition which should not be necessary in any method of chemical analysis. C. J. Schollenberger: Varying the amount of soil used and the conditions of evap- oration in the McIntire procedure affect the results obtained to an extent which prevents this from being a practicable method. If the evaporation of the soil and caleium bicarbonate solution is carried to dryness, higher results are obtained than if evaporated to a thin paste, as called for in the method. The results also vary with the time required for the evaporation, even when it is carried to the same stage. On the other hand, a longer contact between soil and bicarbonate solution before beginning evaporation seems to have but a slight effect, provided the conditions during the subsequent evaporation are not varied. These points are brought out in the table included; the soil used was sample 1. This soil has a rather high lime requirement as determined by this method; a 10-gram sample decomposed somewhat more than one-half the caleium bicarbonate contained in 100 ee. of the solution used; a 20-gram sample, then, should use up the entire quantity. As a matter of fact, only about three-quarters of the total calcium carbonate originally in the solu- tion was decomposed when a 20-gram sample was used. The Marr method was used for the determination of residual calcium carbonate. The same objections will no doubt apply to the Hutchinson-MacLennan method, namely, that the results obtained will depend upon a number of factors. 1917] AMES: SOILS 137 A soil which showed a calcium absorption equivalent to 3,025 pounds calcium carbonate per 2,000,000 as determined by shaking 40 grams soil in 200 ce. solution for 3 hours, gave a figure equivalent to 3,600 pounds calcium carbonate per 2,000,000 when 100 grams of the same soil was shaken with 500 cc. of the same bicarbonate solution for 3 hours, then allowed to stand overnight before filtering and titrating. For purposes of comparison, the lime requirements as determined by several methods on soil No. 1 are given below. Lime requirement as pounds calcium carbonate per 2,000,000 soil. Hopkins-Pettiti methods arcs ae sss clone wise tos eisin ss oa.cie ate ce meee 600 Hutchinson-MacLennan method....................ec0eeeeeeeees 1,280 Mein piresmie LOGE Eee tit aporsieiecs 2. eked oie isin abe sesctooe 7,500 SOW eo UUM YIN e LN OCderyanec\ciaianarsieletavainvsre Gc ciatera eins s/s (eerste are evatejateve'oaNe 6,200 DISCUSSION. The results for lime requirement show a wide divergence for the same soil by the several methods used by those who submitted a report on this subject. By varying the proportion of soil used in the Hutchinson-Mac- Lennan method, a calcium-carbonate requirement of from 400 to 5,800 pounds per acre is indicated. The McIntire method also shows the same sort of variation. Results by this method are also influenced con- siderably by the conditions of evaporation and the stage to which the evaporation of soil and solution is carried. All the results from various sources for different soils contributed to this study of lime requirement methods emphasize the fact that the Hutchinson-MacLennan method and the McIntire method, as well as all others proposed, are empirical in nature and that comparatively slight variations in procedure affect the results markedly. Results obtained by C. J. Schollenberger with the McIntire method show that this method is quite sensitive in this respect. This is true of methods in which the soil is in contact with solution of calcium hydroxid, carbonate, or solid calcium carbonate and water, as it is of methods which employ a salt solution. If another salt is substituted, the concentration of the solu- tion, or the condition of the soil and the proportion of soil to solution, be changed, different results will be obtained. Whatever may be the phenomena which gives rise to the so-called acid condition of soil, it can scarcely be expected that any laboratory method will give an indication of the amount of lime which from an economic point of view will be practical for field conditions, for the reason that the reactions between the lime applied to the soil in practice will not take place under the same conditions or to the same extent as when the soil is prepared for the analytical procedure involved. While methods suggested for the determination of lime requirement are of value in studying the processes which are taking place in the soil, they can not give other than an approximate measure of the soil’s need 138 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 of bases, which is generally indicated in the case of soils deficient in calcium and magnesium carbonates. It does not seem advisable that any method for lime requirement or soil acidity be adopted and included in the official methods of the asso- ciation. RECOMMENDATIONS. It is reeommended— (1) That the following modification of the Marr! method be adopted as provisional for determination of inorganic carbon in soils: Place from 5 to 20 grams of soil, depending upon the carbonate content as indi- cated by a qualitative examination, in a suitable flask or bottle having a capacity of 250 ce. and which will withstand a vacuum of approximately 70cm. Add 80 ce. carbon-dioxid-free water; after mixing thoroughly connect flask to suitable appara- tus? provided with condenser and Meyer absorption tube. Start suction, and when air has been removed from apparatus add to the contents of flask, through separa- tory funnel, 20 ce. dilute hydrochloric acid (2 ce. hydrochloric (specific gravity 1.19) to 18 ce. water).? Allow acid to act on soil for from 5 to 15 minutes, continuing suction, before heat- ing contents of flask. Then boil for from 20 to 30 minutes, maintaining a vacuum of 65 to 70 em. in the boiling flask. Care should be taken that solution in flask is not drawn up into condenser tube. Absorb carbon dioxid evolved in a suitable quantity of from one-third to one-half saturated barium hydroxid solution, con- tained in a Meyer absorption tube. The barium carbonate after filtering and washing can be determined either volumetrically‘ or gravimetrically. A blank determination must be made under same conditions and correction applied. (2) That the referee on:soils for 1916 study methods for total sulphur in soils. It is suggested that a comparison of the following methods be made: Sodium peroxid fusion; heating soil with magnesium nitrate solu- tion as used for total phosphorus in soils; modification of Eschka’s method for sulphur in coal, ignition of soil with mixture of magnesium oxid, sodium carbonate, and ammonium nitrate. (3) That methods for extracting sulphates from soils be studied. (4) That methods for the determination of the total constitutents of soils be studied with a view to substituting them for the “strong acid digestion’’ as outlined under section 5, page 14, U. 8. Bureau of Chem- istry Bulletin 107 (revised). 1J. Agr. Sci., 3: (2) 155. 2 Similar to apparatus described in J. Ind. Eng. Chem. 6: 561, but omitting Camp absorption tower and substituting Meyer absorption tube. ‘This proportion of hydrochloric plus the 80 cc. of water previously added gives a strength of acid for decomposition of carbonates of 2 to 100. If the nature of the soil is such that a greater strength of acid is considered necessary, an amount of acid can be taken to make the strength of acid used for digesting soil 3 to 100. 4J.R. Cain. J. Ind. Eng. Chem., 6: 465. 1917] CONNER: LIME REQUIREMENTS OF SOME ACID SOILS 139 LIME REQUIREMENTS OF SOME ACID SOILS. By S. D. Conner (Agricultural Experiment Station, Lafayette, Ind.). A large proportion of the soils of the eastern half of the United States are more or less acid. These soils vary in type from sands, silts, and clays low in organic matter to peats with 80% to 90% volatile matter. Lime experiments have been conducted in field or pot tests on several types of acid soils as follows: EXPERIMENTS. Soil No. 1, which is a very acid peat, failed to respond in a pot test with corn to an application of calcium carbonate equivalent to 16,000 pounds per acre, while it did respond favorably to an application of 32,000 pounds. - Soil No. 2 is a peat that has not responded profitably to lime treat- ment when corn was grown. The total increase of corn in a four-year field test was 91 bushels per acre with a fertilizer rich in potash. When 1,000 pounds per acre of CaO was used together with the fertilizer, the increase was 89 bushels per acre. Three thousand pounds of CaO used alone gave in three years a total increase of 13 bushels of corn per acre. Soil No. 3, which may be classed as a peaty sand, gave in a three- year field test with eight crops, including corn, wheat, oats, and soy beans, a total yield of 46 bushels of grain and beans per acre when fer- tilizer was used alone. With 2 tons to the acre of ground limestone added to the fertilizer, the total yield was 201 bushels. With 4 tons of limestone, the total yield was 222 bushels, and with 8 tons of limestone added, the yield was 229 bushels. Soil No. 4 is a loam well supplied with organic matter. Lime has been used on this soil without noticeable effect on general crops. Clover variety work has been conducted on this soil without lime for the last three years with satisfactory results. No lime experiments have been conducted on clay soil No. 5. Soil No. 6 is a white silt or silt loam, which responds to lime on all crops tried, both in the field and in pots. It is interesting to note that in pot work with this soil, clover entirely fails to grow on the untreated soil. Ground limestone at the rate of 2 tons to the acre produces good clover. Ground clover chaff in an application one-third as heavy as the limestone has produced an equally good crop of clover, although the soil still remained acid. Four soil-acidity methods have been compared in making laboratory estimations of the lime required for these soils. The accompanying table gives the results of these tests, using Hopkins’s potassium-nitrate method, 140 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Veitch’s limewater method, Hutchinson and MacLennan’s calcium-bicar- bonate method, and Jones’s calcium-acetate method. Results on acidity of various soils. CaCO, NEEDED PER 2,000,000 PouNDS OF SOIL VOLATILE Hutchi la te eae Hopkins Veitch ‘ ard C. H. Jones method method |MacLennan| method method per cent pounds pounds pounds pounds No: 1: (peat) sceateaccene soe ere 83.5 8,000 96,430 49,600 74,060 No: 2) (peat) aeaeancseen tek nero 86.2 1,110 49,290 21,600 36,200 INO! on (DIS ckananG) pee seeeeeeen ce 8.2 3,670 17,350 12,000 14,260 INO 4 (loam) eee ee eres 7.2 320 3,210 5,200 7,130 NosS clay) scheteerecen teens 3.9 6,170 | 6,070 | 6,800 | 9,330 INON16) (Silb) Reece teeta eee CEE 3.0 1,110 1,430 2,400 3,020 DISCUSSION OF RESULTS. It is seen that the Hopkins method gives much lower results than any of the other methods, especially on the soils containing much organic matter. This method possibly gives too low an estimate as an average; but even so, in most cases it is Just as near as any other method to the amounts of lime which have given profitable returns by actual tests on the crops. While the Veitch method doubtless more accurately determines the amount of lime that a soil has the capacity to absorb, it is seen that it is not profitable to add this much lime to soils high in organic matter. The results of these tests indicate that organic acidity is much less toxic in soils than inorganic acidity. The Jones calcium-acetate method gives higher results than the cal- cium-bicarbonate method. If, however, the titration figures of the Jones method were multiplied by the factor 1.35 instead of 1.8 these two methods would give much more accordant results. None of the soil-acidity methods can be used as an exact estimate of the most profitable amount of lime to be added to the soil. It is only when combined with experience and a knowledge of the use of lime as shown in actual field tests that any labo- ratory method is valuable. In such connection, I believe a soil-acidity estimation is probably the most important single test that can be made in the laboratory to determine the chemical requirements of the soil. 1917] HOWARD: DETERMINATION OF LIME REQUIREMENTS 141 DETERMINATION OF THE LIME REQUIREMENTS OF SOILS BY THE USE OF CALCIUM BICARBONATE. By L. P. Howarp (Agricultural Experiment Station, Kingston, R. I.) In connection with one of our ecological problems, it became necessary to have a very accurate method for the determination of soil acidity. The method, above all, whether it bore any relation to agricultural prac- tice, tedious or simple in its technic, must satisfy one condition, namely, it must yield a “requirement” independent, within reasonable limits, of the amount of reacting base and the time of the reaction; that is, it must give the maximum base absorption power of the soil. No method was available that would stand this test, but the oppor- tune appearance of two new methods in the literature was very welcome. In neither case was any data incorporated in the article to show in how far it might be expected to fulfill our requirements, and it was also readily recognized that each might prove entirely satisfactory with a given soil type and the purpose for which it was devised. We decided to give them careful consideration. Both methods made use of a solution of calcium bicarbonate—ideal in its relation to agricultural practice. HUTCHINSON-MacLENNAN METHOD. The first to appear was the Hutchinson-MacLennan' method, and it will be discussed first. Its essential points are as follows: Ten to 20 grams of soil are placed in a bottle of 500 to 1,000 cc. capac- ity, with 200 to 300 cc. of approximately N/50 calcium bicarbonate. The air in the bottle is displaced by carbon dioxid, the bottle tightly stoppered and placed in a shaking machine for three hours. The solution is then filtered, and a portion of the filtrate equal to half the original amount of bicarbonate is titrated with N/10 acid, using methyl orange as the indica- tor. TABLE 1. Relation of the “requirement’’ to the amount of calcium bicarbonate used. aO PER 2, s OCT ncaa Cc 000,000 PouNDS OF SOIL BICARBONATE ADDED Soil No. 5 Soil No. 11 grams ce pounds pounds 10 100 2,745 4,973 10 200 3,672 6,126 10 200 3,745 6,294 10 300 4,177 6,552 10 300 4,095 6,418 10 UG oe Beem Weed aaescaae dace ac 7,067 10 BOOT ae ie Sonia ik 7,414 1 Chem. News, 1914, p. 2854. 142 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 A few trials made at this station were sufficient to show that the reac- tion was complete in three hours and this time was used in the subse- quent work. CONCLUSIONS. (1) The reaction is complete in three hours. (2) The requirement varies according to the amount of reacting base in contact with the soil. (3) With an arbitrary procedure, closely agreeing duplicates are to be had. McINTIRE METHOD. The McIntire! method, as proposed before the association at the meet- ing a year ago, in principle is this: Ten grams of soil are treated with 100 ec. of calcium bicarbonate; the solution quickly evaporated to a thin paste. The residual caleium carbonate is then determined by meas- uring the carbon dioxid, and this amount subtracted from that added represents that absorbed by the soil. The carbonate was completely decomposed by phosphoric acid, with no appreciable decomposition of organic matter; the time of aspiration was 30 minutes—factors asserted in the method and substantiated by us. The carbon dioxid was measured volumetrically; and of the several methods employed, the limit of error was least by proceeding in the following manner: The gas was absorbed in the prescribed manner, the carbonate pre- cipitated with neutral barium chlorid; the solution made neutral to phe- nolphthalein by slowly adding acid; sufficient standard N/20 acid was added to decompose the barium carbonate; the carbon dioxid boiled out; the solution cooled, and the excess acid titrated with N/50 NaOH. TABLE 2. Relation of the ‘‘requirement’’ to the amount of calcium bicarbonate added. CaO PER 2,000,000 PouNDSs OF SOIL VOLUME CALCIUM Lait BICARBONATE ADDED : Soil No. 30 Soil No. 11 Soil No. 15 grams cc. pounds pounds pounds 10 BOO) i cece vs vexenans Buxtevarn evel bere emyeteveleynipte ters 3,528 10 150 10,640 107724 ee 10 } 200 12,684 W25555) 7 Wis. cote creneievceeteiee 10 300 17,052 19,328 | | tidoey eee 10 400 Z0;720R dlniias minerieeero: 15,624 10 500 24.080 —" Fliscionew sem den ace css.05 ae eee 10 SOO!) | estrectepoae ectas Siebel eserctetevobeteiere etscenie’s 21,112 ! Am. Fertilizer, Nov. 28, 1914, 41: (11) 36; J. Ind. Eng. Chem., 1915, 7: 864. 1917] HOWARD: DETERMINATION OF LIME REQUIREMENTS 143 DISCUSSION OF RESULTS. From the above it is seen that the absorption power of the soil for solu- ble lime is practically unlimited in any amount that could be reasonably employed in a working method. TABLE 3. Relation of the ‘‘requirement’’ to the amount of soil used. sons No.2 Seren aaa grams ce. pounds 10 100 3,797 20 100 3,097 10 : 150 6,205 20 150 3,836 In some correspondence with Mr. McIntire we learned that working with the Tennessee soils a maximum absorption is secured; that is, the addition of a second portion of bicarbonate resulted in no further absorp- tion. With the majority of soils this may hold true, but with ours the procedure is not applicable. Can we not adopt an arbitrary procedure yielding comparative results? With fifteen determinations on a given soil with such a procedure, namely, 10 grams soil and 100 ec. bicarbonate, yielding requirement of 2,820, the probable error of a given determination was + 62.6. For our work such a procedure would be undesirable; for as the absorp- tion depends on the excess of reagent, then with widely different soils several determinations would be necessary to discover the proper volume of bicarbonate to add in order that, at the end of the reaction, the amount of residual calcium carbonate might be identical in each case. It seems that a bona fide correlation would be impossible without these conditions being met. On account of the slight solubility of calcium carbonate in carbonated water and the impossibility of adjusting this excess reagent without increasing the volume of solution, thereby increasing the period of con- tact, a more complete reaction with the silicates and a subsequent con- dition vastly different from that which we endeavor to create results. We tried to correct this condition by adopting a volume of 100 ce. bicarbonate as our standard, thus regulating the time of reaction, and by varying the amount of soil taken. It seemed undesirable to use less than 10 grams of soil, and work with this variation resulted in degrees of absorption which were far from comparable. We tried using purified precipitated calcium carbonate, in a quantity greater than was possible by using a workable volume of the bicarbonate, adding carbon-dioxid water and evaporating. This proved unsatisfac- 144 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 tory in spite of the fact that we learned that on the Tenenssee soils the absorption was proven to take place after the deposition of the calcium carbonate, and the reason for adopting the soluble carbonate was simply to secure a more accurate aliquot. We believe that it is doubtful whether an absolute method can be evolved which is based on the principle of allowing a certain indefinite amount of reacting base to be in contact with the soil at the end of the reaction. We are now working toward this end, to see whether a work- able method can be devised so that at the end of the reaction no excess base remains in contact with the soil. We trust that for those who desire a critical method, our experiences with these methods may be of value. STATUS OF THE PROBLEM OF LIME REQUIREMENT. By W. H. McIntirz (Agricultural Experiment Station, Knoxville, Tenn.). The study of the intricate problem of soil acidity or lime requirement has now reached a point where it is well to review and definitely classify the work so far accomplished. But more particularly is it desirable that we designate what constitutes soil acidity or lime requirement, and that there be adopted a terminology which will convey a definite and an ac- cepted meaning. Methods for the determination of soil acidity or lime requirement have originated from various lines of thought, the consideration of feasible technic often being the deciding factor in advancing a procedure. In some of the work upon soil acidity, much has been assumed, and often qualitative absorption reactions have been used as quantitative acidity indications. The interpretation of the results thus obtained is based upon the assumption that relatively insoluble carbonate of lime would be utilized by the soil in amounts chemically equivalent to the amounts of the CO.-free, water-soluble alkali salts with which a soil might be treated. The fallacy of such an assumption may easily be shown. First, no close correlation between a soil’s reaction with Na,O and K:O can be shown, nor can any correlation between CaO and MgO soil reactions be expressed chemically. Again, during a given time, the absorption will vary with the strength of the soluble salt solution. Let us consider for a moment the relation of the activities between soils and CaCO; and MegCOs, the alkali earths whose properties and reactions we most com- monly consider as being comparable. There is a vast difference between the ability of a given soil to react toward one of these carbonates and that which it exhibits toward the other. Many figures could be cited in 1917] M’INTIRE: STATUS OF PROBLEM OF LIME REQUIREMENT 145 proof of the statement, but the following may be given. In eight instances a given soil was treated under field conditions with 32,000 pounds equiva- lent of CaCO; per 2,000,000 pounds of soil and checked against CaO, Ca(OH)2, and CaCO; in chemical equivalence. Without leaching, the MgCoO; had entirely disappeared in eight weeks, while from each of the three forms of lime about 20,000 pounds of CaCO; remained. Again, treatments of CaCO; and of MgCOs; in equivalence of 32 tons of CaO per 2,000,000 pounds of soil gave a difference in residual carbonates of 1.67%, or 33,400 pounds on the original soil basis, while equivalent appli- cations of the two carbonates in amounts equivalent to 200,000 pounds of CaO gave 2.82%, or 56,400 pounds more of residual CaCO; than of MgCoO;. These latter data were obtained in lysimeter investigations one year after treatment. Furthermore, a previous excessive treatment of MgCO; upon an acid soil will inhibit its ability to further effect an immediate decomposition of CaCO;; however, the presence of an excess of CaCOs, after satisfying lime requirement, will not preclude an exten- sive decomposition of MgCO; by the then alkaline soil. Again, strange as it may seem, some soils previously treated with both CaCO; and Na2CO; still possess to a marked degree the ability to decompose MgCOs, when this is added in carbonated-water solution and thrown out in con- tact with the soil by evaporation. In other words, there is no relation between the soil’s reaction toward finely divided CaCO; and finely divided MgCO;. Excessive, indeed, would have been lime-requirement results if by chance the more soluble precipitated MgCO; had been used in carbonated-water solution and the results thus obtained computed to terms of CaO, upon the assumption of chemical equivalence in reaction. Such a procedure would, however, have insured in every instance a sufficient lime-requirement indication. The more recent trend of opinion has been toward the belief that the correct laboratory procedure to follow in determining soil acidity is to utilize the same material which is used in the field. The idea most prevalent seems to be that if correlation be sought between laboratory methods and field requirements, a soil should have an opportunity to combine with the same compound under both laboratory and field con- ditions. Since it has been shown that CaO and Ca(OH). as used in practice quickly change to carbonate, the latter substance becomes the logical one to use in laboratory practice. But, even an agreement upon this point is not sufficiently definite. Not only must the form of lime be constant, but the fineness and purity of the material must be uniform. There is, both in field and in laboratory, an appreciable difference between the extent of the reaction of an acid soil with precipitated CaCO, and its activity toward finely ground limestone of comparable purity. If an acid soil be evaporated separately with equal amounts of precipitated JOUR. ASSOC. OFFICIAL AGRIC. CHEMISTS, VOL. III, NO. 1 146 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 chalk and very finely ground limestone, the decomposition of CaCO; effected by the soil’s acidity is greater in the case of the precipitated chalk than in the case of the limestone. However, in field treatment there is a close parallel to be found between the neutralizing activity of CaCOs;, formed in the soil from applied burnt lime and from hydrated lime, and that exerted by precipitated chalk. Moreover, in the parallel of the more active precipitated MgCO;, we have data tending to show that, when applied in excessive amounts, the difference in rapidity of decomposition of a “fluffy”? precipitate by an acid soil may be quantita- tively denoted as being greater than that of the more crystalline pre- cipitated MgCO3. Again, in addition to variations in degree of fineness, precipitated chalk almost invariably carries some hydrates which must be carbonated by carbonated-water treatment, if this material is to be used for quantitative lime-requirement work. There is but one feasible method of insuring universal uniformity of condition in treating soil with CaCO;, and that is to apply the carbonate as a solution in car- bonated water. In so far as we are aware, no stress has been placed upon this fact. For a period of three years the factors influencing the lime require- ments of soils have been studied by the Tennessee Agricultural Experiment Station, as an Adams fund project. When the work was begun in 1912, there were, for the determination of lime requirement, but three methods which directed the use of oxid or carbonate of lime, to wit, the Veitch, the Tache, and the Siichting methods. Of these, only the Veitch gave, after treatment as directed, a soil residue which would produce in the laboratory no further lime requirement immediately subsequent to the acidity test. In other words, of the three methods, only the Veitch gave opportunity for maximum decomposition. It seemed safe to assume that methods which will give but partial indication of a maxi- mum possibility will vary in their partial results, with varying laboratory conditions. The Veitch method was therefore used as a basis of study, with the hope that its technic might be so modified as to eliminate its laborious features and to render it both rapid and accurate. The laboratory inves- tigations were supplemented by the use of 440 baskets containing three distinct soil types. After trying various modification schemes we found it impracticable to modify the Veitch method for quantitative estima- tions. We found, however, that upon taking up with distilled water after evaporation, the soil mixture could be agitated and thrown im- mediately upon a 10-em. Biichner funnel, thereby securing quickly a clearer filtrate than the supernatant solution to be had after stand- ing overnight. An effort was then made to secure a procedure which would fulfill the requirements of the Veitch method and serve as a 1917] += M’INTIRE: STATUS OF PROBLEM OF LIME REQUIREMENT 147 substitute for it. It must be borne in mind that in such a study a certain conception of lime requirement must be held. Upon this point, however, there appears to be no unanimity of opinion. There is no official, provisionally accepted, or adopted definition of the term “soil acidity” or “lime requirement.’’ Hence, it was necessary first to assume certain conditions and to evolve a method which would meet these con- ditions. The assumption followed was that the lime requirement of a soil may be designated as its ability to decompose a maximum amount . of CaCO; under laboratory conditions, without the decomposition of any neutral organic matter. Such a hypothesis is in harmony with that tacitly assumed in the Veitch procedure. As before stated, from such a study it was found that the only uniform quantitative pro- cedure dependent upon the estimation of residual carbonates by CO: determinations involved the use of a CaCO; solution. Evaporation of soil and CaCO; effected a greater decomposition of CaCO; than did boiling for a 5-minute period, and apparently resulted in no decomposition of neutral organic matter. The contact of a partially saturated car- bonated solution and soil need be of but brief duration, and no secondary hydrolysis of native minerals is then effected, because of the evaporation which expels the gaseous CO2. Conditions fulfilling those of the Veitch method were thus met. The work was carried beyond this point, however, and it was found that after more than satisfying the lime requirement according to the Veitch procedure, unleached and sterile soils were still able to effect, by ordinary contact, continued and appreciable decompositions of CaCQ; with fixation of CaO, principally in the form of silicates. This is in har- mony with the work of Veitch, Gardner and Brown, and Hutchinson and MacLennan, who showed that the satisfaction of a soil’s original lime requirement, according to indications of the Veitch procedure, does not preclude an additional lime requirement upon unleached potted soils. Elimination of the question of organic matter and biological factors estab- lished the fact that the continued decomposition of CaCO; must be due either to belated reactions of certain less active acid silicates or to hy- drated SiO., or to a combination of the two causes. Thus we were led to the conclusion that, speaking chemically, it is necessary to differentiate between what might be termed temporary, immediate, or apparent laboratory lime requirement and continued lime requirement of soils. Contrary to some beliefs, in the rare cases in practice where excess of lime is applied, any excess does not remain inert, save for the loss through leaching, interchange with bases, neutralization of nitric and possibly other acids. In other words, where a soil contains an excess of CaCOs, the excess is diminished gradually, net alone by neutralization, leaching, 148 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 1 and replacement of alkalis, but also by a continued reaction not only with silicates but also with hydrated SiO. and TiO. The combined influ- ence of these several factors upon the aggregate residual CaO derived from applied CaCO; has been shown to be of considerable importance in practice. Analyses of the twelve limed plats of the Pennsylvania Agri- cultural Experiment Station after 32 years of liming showed that of the residual CaO above that of the check plats, about 35% is to be found in silicate combinations. The soil of the check plots, today but slightly acid, and in some cases either neutral or alkaline, has, therefore, accom- plished an extensive decomposition of CaCO and the fixation of the lime in the soil in combination with silicic acid. This brings us to the con- sideration of the practical value of a lime requirement determination, and to the point of emphasizing the need of some fixed standard to be met by such a determination. By lime requirement, do we mean a soil’s partial or its maximum ability to fix CaO through decomposition of applied lime compounds in the laboratory alone, or do we intend that the laboratory practice shall approach at least a correlation to crop response under field practice? Is it our aim to use a method to determine the laboratory condition of a soil, or to secure data which will afford an indication of its actual or approximate field needs? Would not the latter involve extensive field work over a period of years? Certainly pot or basket work is not sufficient in scope to be altogether conclusive. It is our belief that in selecting a method for the determination of such an indefinite and elusive soil property as lime requirement, there is need for research, rather than mere comparisons of methods. In codper- ative work upon soil problems such as acidity, the referee is confronted with difficulties which do not hinder the referee upon those subjects where various proposed methods may be compared as to their accuracy upon synthetic solutions. Where there is no unanimity of opinion wpon what actually constitutes a desired reaction, he is at a loss to decide when conditions sought are attained, or which of two or more sets of results may be correct. While affording opportunity for extensive comparisons between methods, our referee system permits of no research to establish the authenticity of results obtained, so far as soils are concerned. Quite naturally, this results in a tendency to lean toward a procedure which offers easy or attractive technic. A case in point is the study of the two methods of lime requirement during the current year. From a research of the fac- tors governing soil acidity, we know that one gives an alkaline condition to the soil residues when these are treated in blank according to the Veitch procedure, while the other falls short of the Veitch requirement, and its residue will give a further lime requirement by duplication of its own procedure, or by following any one of four other methods. One repre- sents a maximum reaction, the other a partial one, and one dependent 1917] NOYES: DETERMINATION OF PHOSPHORUS IN SOILS 149 upon and varying with the amount of charge. But, which indication is correct, the partial or the complete? What is the viewpoint to be as- sumed, that of the laboratory or that of the field? If field applications be taken as the viewpoint, how are we to know which procedure may be preferable? It is quite possible that two methods would give indications which might give closely concordant increases in yields for the first year, but after that, returns which might indicate the advisa- bility of the larger indication. Furthermore, recent work at thelowa Agricultural Experiment Station has shown that fineness of division of the sample taken for analyses is an important consideration. That is, a soil, acid when coarsely ground, may become alkaline to Veitch tests when more finely. ground. More recently the reverse has also been shown to be possible. What, then should be the fineness of the analyzed sample? This association has shown its appreciation of the importance of the phosphate question by appointing a very able committee to study this problem. It will be readily conceded that in mary cases liming is a pre- requisite to the profitable use of acid phosphate as well as other fertil- izers and that the subject of lime is, therefore, of equal importance to that of phosphate availability. Would it not, therefore, be in line of conservative yet progressive accomplishment for this association to desig- nate a committee of its members to definitely determine what shall be officially considered as constituting lime requirement, to decide upon a definite degree of fineness, and to further decide whether it be advisable for this association to attempt a correlation between laboratory procedure and field practice, or whether the laboratory problem alone shall be solved, leaving the field problem of the variable factors of crop and influ- ence of soil type and climatic conditions to such an organization as the American Society of Agronomy? In the treatment of the lime-requirement problem, such questions could and would be carefully considered and settled, were this association to adopt toward it a policy similar to its method of handling the phosphate problem. But at least, let us cease to evade the issue; let us be agreed as to what constitutes immediate, temporary, or apparent soil acidity and adopt for such some definite terminology. DETERMINATION OF PHOSPHORUS IN SOILS. By H. A. Noyes (Agricultural Experiment Station, Lafayette, Ind.). Due to evidence presented by Hilgard,! Goss,? and others, it is well established that strong acids will dissolve the phosphorus present in soils. 1U. S. Dept. Agr., Div. Chem. Bul. 38. 2U. 8S. Dept. Agr., Div. Chem. Bul. 43. 150 ASSOCIATION OF OFFICIAL AGRICULTURAL cHEMisTs [Vol. III, No. 1 Since the method of Goss dissolves the phosphorus and requires less time to prepare a solution for analysis, it has been used in all investi- gational work conducted by the author of this paper. With clay soils, subsoils, and some freak soils difficulty is experienced in dissolving with nitric acid the precipitate formed on preliminary neutralization of the prepared solutions. It has also been observed that those samples where it is hard to dissolve the ammonium hydroxid precipitate are the ones where later on in the procedure the yellow precipitate is hard to dissolve and where clean, clear solutions are not present when the phosphorus is reprecipitated as magnesium ammonium phosphate. Sulphuric acid is a weaker acid than nitric acid. This leads us to believe that if a proper amount of ammonium nitrate was added to the original aliquot there would be no loss of phosphorus due to increased solubility of ammonium phosphomolybdate. When the laws of physical and chemical equilibrium are applied to solutions that have and have not been neutralized before precipitation of ammonium phosphomolybdate we find the same possibilities of ionic concentration, molecular combination, and solubility. This is true if ammonium nitrate is added to the solutions. That the positive H ions in the solution in either case are offset principally by negative NO; ions is borne out by the fact that nitric is a stronger acid than sulphuric and ammonium sulphate is less soluble than ammonium nitrate. The P.O; results given in the following table were obtained on 10 soils chosen for their variation in acid oxidizable material. The method of analysis is that of Goss,! except that for every 25 ce. aliquot of solution taken 15 grams of dry ammonium nitrate is added at the outset to the set of solutions that received no preliminary neutralization. Soil analysis—dry basis. P20; NEU- P20s NG INORGANIC ORGANIC OH TRALIZED Na COz Cc N per cent per cent per cent per cent per cent Lronisoilii.-8 229 re Le 1.50 0.068 4.38 0.64 I BYoyRieeae asp aces Hou Gos *2.45 2.41 0.018 0.75 0.13 Black-soilivn) see ereencers 0.53 0.53 0.033 18.63 2.41 Acidipentay: santa eee 0.44 0.45 0.002 23.88 2.16 Blackjsandsssse eee 0.41 0.43 0.018 2.16 0.38 York silt loam....... se 0.12 0.13 0.000 0.51 0.02 Gumbott.3.5 See 0.47 0.48 0.230 Lill, 0.41 Hog sloame.. perpetereee 0.34 0.34 0.101 0.95 0.16 Red sand....... ; *0.16 0.15 0.000 0.38 0.05 Decatur clay loam...........| *0.17 0.15 0.030 0.59 0.09 The results indicated by an asterisk (*) in the table denote those solu- tions which on neutralization with ammonia did not give precipitates ‘Wiley. Principles and Practice of Agricultural Analysis, vol. 1, p. 465. 1917] NOYES: SOIL CONTAINING RESIDUAL LIMESTONE 151 that were readily redissolved by nitric acid and which subsequently gave slightly cloudy solutions on standing with the magnesium ammonium phosphate precipitate. CONCLUSION. Nothing is gained by the neutralization of the phosphorus solution with ammonia and subsequent dissolving of precipitate with nitric acid before precipitation of ammonium phosphomolybdate. Neutralization may precipitate substances from the original solution that are not easily redissolved after being precipitated. STUDY OF SOIL CONTAINING RESIDUAL LIMESTONE. By H. A. Noyes (Agricultural Experiment Station, Lafayette, Ind.). In connection with an Adams-fund project on orchard management, under investigation in the horticultural department, the acidity of the soil at different depths was studied. The soil is a residual silty clay containing about 60% very fine silt and 20% clay, underlain with lime- stone rock. Samples were taken representing the following depths: 0 to 9 inches, 9 to 18 inches, 18 to 27 inches, 27 to 36 inches, and 36 to 45 inches, where no bed rock was found above the 45-inch depth.. The samples representing the different depths were air-dried, fined with a wooden rolling pin, and sieved through a 40-mesh sieve. Limestone fragments varying in size from that of a kernel of wheat to 2 cm. in diameter were found in eight samples representing three of the places chosen for sampling. The following table gives the results of determina- tions made on the samples taken at these three places. Stress is not laid on the amount of acidity, nor on the accuracy of the method employed for determining acidity, but on the fact that this standard method places the soil as acid. The acidity, determined by the Hopkins potassium- nitrate method, is expressed as pounds of calcium carbonate necessary to neutralize 9-inch layers (3,000,000 pounds) of this soil. The column headed calcium carbonate reports the acid neutralizing powers of the limestone particles where found in the soil as percentage purity of cal- cium carbonate. Columns headed hygroscopic moisture, volatile matter, and nitrogen are given as per cent of air-dry soil. 152 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Determinations on residual limestone soils. ACIDITY EUBILY d SAMPLE NO. DEPTH gees S55 pred OEE NITROGEN CaCOs Sa MOISTURE tnches pounds per cent per cent per cent | per cent 2B ah penteaterone Oto $ LPIA seco 1.96 4.39 0.12 9 to 18 O19 29 etree 2.50 4.74 0.11 18 to 27 Uhey (Gy | Reena (4) (9) (1) 27 to 36 31) 0)e Beainoere (4) (2) @) 36 to 41 (@) 93.8 3.85 0.06 b,@ LOR Pench aa aed Oto 9 TOBE S| ersie eps oe 3.31 4.75 0.10 9 to 18 UST2 Dal teers 3.66 5.43 0.09 18 to 27 37.5 | 73.8 2.70 6.94 0.06 27 to 36 56.4 | 89.9 2.73 6.46 0.06 36 to 45 150.0 | 89.0 3.80 5.52 0.07 TER 5 1) cc aria Oto 9 TP BGYe Iecrdo cota 2.04 6.49 0.21 9 to 18 (?) 94.3 2.63 5.75 0.13 18 to 27 56.4 | 96.3 2.97 5.43 0.10 27 to 36 75.0} 90.4 2.40 Catt 0.11 36 to 40 18.9} 83.5 2.27 8.68 0.05 1 Not determined. 2 Sweet. DISCUSSION. The acidity does not decrease the deeper we go down into the soil. Tn all three places reported upon here, as well as in the other places investi- gated, the acidity shows no regular decrease or increase dependent upon the distance from bed rock. The acidity results can not be correlated either with the percentage of nitrogen or volatile matter in the samples. The volatile-matter figures probably are governed more by variations in combined water than by actual organic-matter content. The state of fineness of limestone for agricultural use in correcting soil acidity has been a subject of both experimentation and comment. In the 1912-138 report of the Pennsylvania Agricultural College, page 214, we note the following: ‘‘ We conclude, therefore, that on silty loams and on soils of heavier texture, on lands where soil acidity is the chief factor limiting clover production, crushed limestone used for amendment should be at least 60-mesh in fineness of pulverization.”’ This statement is based on experimental results using limestone of varying fineness and clover as the indicator of its effect in correcting acidity. The presence of limestone fragments in the residual acid soil reported upon here points to the conclusion that coarse screenings would not cor- rect the acidity. The Pennsylvania report shows results of lime when it is applied experi- mentally. The soil reported on here has coarse particles of limestone left from nature’s original supply, and yet it is acid. 1917] PATTEN: INORGANIC PLANT CONSTITUENTS 153 It is taken for granted that acid loams, silts, and clays, of which there are large areas, need clover and that clover responds better on a soil neutral or alkaline in reaction. In conclusion, silt and clay soils need finely ground limestone and not coarse screenings to correct soil acidity. REPORT ON INORGANIC PLANT CONSTITUENTS. By A. J. Parren (Agricultural Experiment Station, East Lansing, Mich.), Referee. In presenting a report on this subject, I wish to briefly review the work carried on during the past four or five years, in so far as it relates to the discussion which follows. In 1919 the molybdate method for the separation of ferric and aluminic oxids was studied. This method provides for the removal of the phos- phoric acid by precipitating with the usual molybdate solution, and the precipitation of the iron and aluminum in the filtrate, by cautiously making it ammoniacal, keeping the temperature below 40°C. For this work a synthetic solution containing 30% CaO, 10% MgO, 2% Fe:Os, 3.98% Al:O3, and 9.70% P2Os was used. In 1911, on a solution containing 23% CaO, 10% MgO, 3% Fes, 2% Al.Oz, and 16.81% P.Os;, the molybdate method was again studied, and in addition the oxalate method for the separation of ferric and alu- minic oxids. Some results were also reported by the referee on an exten- sion of the molybdate method to include calcium and magnesium. In 1912 the same methods were again studied, but no results were reported. However, the molybdate method for the separation of ferric and aluminic oxids was made official, and it was recommended that further work on the oxalate method be discontinued and that the molyb- date method, extended to include calcium and magnesium, be further studied. In 1913 on a solution containing 11.38% CaO and 9.30% MgO the ex- tended molybdate method was further studied and the following recom- mendation was made: That the official method for iron and aluminum be extended to include calcium and magnesium in the presence of minute quantities of manganese. No report was presented in 1914, and the same recommendation was referred to the present referee. The work carried on during the past year has, therefore, been confined to a study of methods for the determi- nation of calcium and magnesium. The reasons advanced for extending the molybdate method to include calcium and magnesium may, perhaps, 154 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. ITI, No. 1 be best expressed by quoting from the report of a former referee, under whose direction most of the work has been carried on. It is not the intention of the referee to have the new scheme of analysis take the place of the present one, but rather to supplement it, if after it has been tested by the association it is found satisfactory, as there is no doubt that there will be much time and work saved by using it. The method will have another advantage in that the phosphoric acid, ferric and aluminic oxids, calcium and magnesium oxids, and possibly the manganese can be estimated in the same solution on one-half gram of ash, which is very desirable when the sample is small. Still another advantage which might be mentioned in its favor consists in avoiding the acetate separation, which at best is not very satisfactory in the hands of the average analyst. On reviewing the results reported in previous years for calcium and magnesium by this method, they are found, in most instances, to be very satisfactory except where manganese was present in the ash solution. In such cases the calcium precipitate was always contaminated with manganese, which, where the amount was determined, was found to vary from 0.3 to 2 mg. Mn,Qx. Comparisons have been made in the referee’s laboratory between the official and molybdate methods for the determination of calcium and magnesium in solutions approximately the same as those used in other years. The results were practically the same by both methods, except that manganese, when present, was invariably occluded by the calcium precipitate in the molybdate method. The amount was small, however, in every case amounting to less than 4% of the total weight of the calcium oxid. The presence of manganese is plainly visible after ignition, on account of the color imparted to the calcium oxid, and because of this fact the error often appears much greater than is actually the case. The official method provides for the removal of the manganese before the precipitation of calcium, and it is consequently not a source of error in the latter determination. Another disadvantage in the molybdate method is the time required to complete the determinations, and when only calcium and magnesium are to be determined the time factor is a serious drawback. However, in defense of the molybdate method it should be stated that when phos- phorie acid, iron, and aluminum are also to be determined the time fac- tor is no longer an objection. Since all of the previous work on the ex- tension of the molybdate method to include calcium and magnesium has been done on synthetic solutions, it will be of interest to compare them with the ash of a variety of plant materials as given in the following table. No figures were available for ferric and aluminic oxids, but it is fair to assume that they would in every case be less than that for caleium oxid. 1917] PATTEN: INORGANIC PLANT CONSTITUENTS 155 Composition of the ash of plants. PLANT K20 Na2O CaO MgO P20s per cent per cent per cent per cent per cent Seeds of cereals: HOR beer ane ym cern eisies ces 31.0 2.0 33583 12.0 47.3 Bacleyereteer oo. ce = eer 21.3 2.4 2.6 8.7 34.7 OST, 5. an ooo See 17.8 1.6 3.5 de: 25:5 Worms aceeersss ss 29.5 0.9 P00 15.5 45.5 Seeds of legumes Oy? lett 5 sues een se qodor 42.4 0.9 5.5 8.3 36.9 [EGR al oon 43.8 0.5 4.9 8.8 36.0 Rediclovers. 2.525 ses 35.6 ileal 6.6 13.2 38.0 Garden bean............ 41.9 1.0 5.0 72 39.3 Straw: Wheat. 13.7 1.4 5.8 2:5 4.9 iil, gsoren aes sanders 23.3 3.5 7.2 2.7 4.2 Osteria occ cence sss 26.4 3.3 7.0 Bar / 4.6 Warnes. .arccee eects cers 30.1 1.2 10.8 5.6 8.1 SCARS E Sone esisicos tees = 43.2 107, 26.6 5.8 6.4 Hay: Meadow hay............. 25.1 4.5 15.8 7.1 6.1 Rediclover: eee ee F. L. Elliott, Washington, D. C........ 31.90 31.93 BY AA LS Teel ISS nao eae 31.90 31.80 3220355 eee 31.85 SIESOID |e opvae Soe: | Ae eee PAM OESL BG salar shalo sisi spats t cere ees icy s, Voce «(ies 31.88 31.84 32:03 lave J. J. T. Graham, Washington, D. C....| 31.95 32.00 31283).i|saaneneee 31.80 31.90 31590) heaters eee 31.83 32.05 3198" Ea eee Berar ee ence See eee | ai. 86 |) eee, | ato8 |e ae W. J. Morgan, Washington, D. C....... 31.55 31.73 32203) > | Faas see 31.45 31.63 32510) ||Boeee ae 31.50 BIEBSe Iasi cet ese eee CIGD G3). eRe gelpe oe Gaeeeeneion Ca seaoe 31.50 31.75 BI ol eisae oA E. J. Nealon, Washington, D. C........ 31.75 31.70 S21099 Pee eke 31.7. 31.65 32204 tee eons ee See 31.65 S2t0f alee ee ees Beate core 31.55 SPA fant |S kee aan BRMGUACOR SS cea S Ay Sats 4 Sw/deisielvteld stro seie 31.74 31.64 GrANE/, 33 Soeacmee 166 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Analyses of lead arsenate—Continued. TOTAL ARSENIC AS As205 ANALYST Method I (a)|Method I (b)| Method II | Distillation per cent per cent per cent per cent C. H. Walker, Washington, D. C....... 32.11 32.08 32:12) 1). a eee 32.11 32.08 32.16 \i2 eee AV.OT APC: caste sion streeis ra eie co « ee 32.11 32.08 32.14. || Une ote R. C. Roark, Washington, D. C........ 31.90 32.00 32.10 32.00 : 31.80 31.95 32.03 32.00 31.85 31.85 $2.10) |< eee AV OTAGO hai hectonts ns erektrec Ss ol coe nie 31.85 31.93 32.08 32.00 Dean C. Kellog, East Lansing, Mich...| 32.65 32) 60k (eaeees 32.70 32.60 327 80) see eeeee 32.70 AV ETA SEL h voyste tenes cle leetrota tes ertersmen cree 32.63 32.510) || ees eee 32.70 George E. Holm, St. Paul, Minn....... 32.50 32.90 32.20:) |: Race 32.45 33.12 32.26: | .nseeeeee JAVELAR Cost caeie ces wearers «avec cn elelee tes 32.48 33.01 32.23" centres W. H. Rogers and E. R. Tobey, Orono, SL GOs ligheneias 31.97) | Aaa Me. SHAGGY eG 33 ois Saoe 32:05; |-cxeeeeee Chie (\ ae eee eee! EOE Alsrosok sc. - 2) hy ae eerie Meera ces 6 Gcisac> sot AV CYA GC arate rasaieies shefe eis) sie, oh ks Bete ere BLeG8) Tc. cease 32:01 |.iaseeasre iE. E. Sawyer, Orono, Me... . 22.5. .2-5: SiC Gy lee aed ae = 32.42) ”\ epee ene SUAS ieee hati 32.62) |os seers ASV CY BID Ort coats eer ots ote eae ee LAD walle es 32.52) ||..ch eee General ‘average syj.-cr- cose ene 31.92 32.13 32.22 32.28 Methods I (a) and I (b) for total arsenic in lead arsenate are slight modifications of the present official method (U. S. Bur. Chem. Bul. 107 (rev.), p. 237). Method II is here presented to the association for the first time. It is based on the following reaction, which was first studied by Naylor (Pharm. J. and Trans., 1879, (3), 10: 441-442): AsoOs + aunt AsoOs + Al. COMMENTS BY ANALYSTS. W. L. Latshaw and J.C. Ripperton: In Methods I (a) and 1 (b) it was found that heating with sulphuric acid in a casserole for two hours, instead of heating until white fumes were evolved, increased the results 0.5% or more. A. L. Sherman (speaking of Method II for the determination of arsenic as As:Og5 only): I found it was very important to obey all minute precautions throughout this determination. Temperature is an extremely important factor. i. | ; 1917] ROARK: INSECTICIDES 167 In the following table are presented results by the present official, the distillation, and the thiosulphate (II) methods on a number of com- mercial lead arsenates, as well as on some lead arsenates prepared in the Insecticide and Fungicide Laboratory. Results on various lead arsenates. LABORATORY NO. 11548 11549 11550 11552 14604 17730 20076 20232 20350 TOTAL ARSENIC AS As2Os DESCRIPTION a aie = ofissh [Pa | weuca per cent per cent per cent Wommercial’sample: .- 65.0. 0c. en 29019). || Pe eee 29.10 29 Lb wie oesecte. 29.15 PSV CLEP Or sectelcleiasetactete sists vis leleis enetoee 258 Y (0d ers Hecke ie 29.13 @onimniercralisamplet ems - iene 25.76 25.65 2bE2o 25.69 25.60 25.25 LATENT Sap pods oaeeanob en OeoREaoD 25.73 25.63 25.25 @ommercial/samples. ...22....0---- 28 .42 28.70 28.80 28.21 28 .60 28.70 ASV CAPES taiz act fotos fe Aa posi o| toys) dc oh) os 28.32 28.65 28.75 Commercial sample................... 30.44 30.50 31.10 30.73 30.50 31.15 OR gIOoeO bomcceneaS 31.23 JAN GREGG) ods coun» DIB ROE En eee erc 30.59 30.50 31.16 Commercialisumples. cms ec. secs cer DAS |r. cer ree 27.70 De OOW | Ane. aa vee 27.75 AV ETAL CMe eet seesdtacdnaesca ss a feat OS | OF eee ee 27.73 Commercialisample: 2. 3 ....060---- SOLOO) lata eeeciet 30.40 S000) |istcasooous 30.38 PAVERS Pe teme ere ciicdeses.s oe sats haees ase SOCOM ES. aes 30.39 Commercial sample................... leas |e nae 31.50 Cot Ne ay ON ie ane 31.50 IASY CY AE Grr aee ere eee eat ioies aitiaieas ee SIESSH lle eve a: 31.50 Commerciallsamples= >. .-..0-- 5-2-6 7ASaT I emer oF 28.93 28 OTR) Ske. shtece 28.98 AV CTAB ONS isc se decce Sekar Reeser ZeUGOMe Eee See ee 28.96 Commercialisamples-7eeee eee 28-26 | ayaa: 28.33 Posya hci lacoste & 28.35 FAG ea RES GOI OR TAB R TIO SET ae 28/-22,. || Reeser cee 28.34 Commercial/sample..... 2... -200<000n- 302930 |panceteaee 30.95 SIEO5) | Reereraae 30.95 RrcrapeeMna hiya f Rode hace Ue Pil eee | 30.95 168 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMIsTS [Vol. III, No. 1 Results of various lead arsenate—Continued. METHOD Srinaeee cag ea oe Official Distillati method meebo Method IL per cent per cent per cent 20479 Commercial sample..................- BPAY (a lq annta senna 32.60 BZC2O Wane ctetaacets 32.55 LAV ETS OC 'o shisha. = Nooatceeute eaten ns 2588! ) lest wee s 32.58 20832=A:\|"Commercial sample vac see serio a «| (ta = ele ete 29.30 29.65 Ss. pe 29.10 29.55 AV ORS DOLE P98 2:5, Stace ctereeivmerars ee miscnteis| |e. noi cnsteat eae 29.20 29.60 90832-B ii); Commercialisamplet.a..-cccees eter eee less. = ests 29.80 29.55 Fo preeeeee 29.90 29.45 PAVOL REO ae wins.» beaeaemim erie cio as iaie rel] tele cremteletes 29.85 29.50 20832-C)|| (Commercialisammple 2. ssc). sien erste ferns crete ais 29.75 29.65 Soaaeeeels 29.70 29.70 MASVETA GE. ehosiaisis cissispatctieastemions ec clee| ee ees eaars 29.73 29.68 20832=D)) \Gommercialiisamples..-jemr tie cieist>< elses ls si 29.50 29.75 Sareea 29.40 29.80 PAVOTAR ON. casera isto sreieielein/eie ie esate wierezel 4) olstaxcie Meier 29.45 29.78 21165 Commercialisample:..-ss2-% eee eee 31.63 31.70 32.30 31.46 31.60 32.30 Veciodns Joe] Bete 32.25 JASVETIABOR.: Ssjeisic.sts wie - 33.14 33.10 i iaiate fekereiats 33.10 33.15 AV CES D ORR Ren P sl Bris, Seomet ds a ersyeVeh crolaes. | evaracste Cazares 33.12 33.13 Dilead arsenate (PbHAsO,)?........... 33.08 33.08 33.18 33.13 33.05 33.18 Sas batons eeee te cae 33.15 AV ETAP Crm ren eh py atime tie aisreversie eres 33.11 33.07 33.17 Basic: leadiarsenate®: 2.6.0... cm. 23.63 23.59 23.70 23.65 23.56 23.69 PASTETA ZG Mesercses netfee ees at oti oharsiate s 23.59 23.58 23.70 Lead chlor arsenate?................... 23.63 23.63 23.64 23 63 23.60 23.62 Hest, At atc Menem 23.61 JAR GETS ae dag Soo oe ane DCE Ear 23.63 23.62 23.62 1909 A. O. A. C. lead arsenate ......... 30.93 30.86 30.90 30.78 30.85 30.93 SOSOS Ti a2 oat asee eee SOETS:» || ssecav sida rarcllh cc sete efarcters EES STET ORY. ore SEO PEE RCO 830.86 30.85 30.92 1910 A. O. A. C. lead arsenate ......... 31.98 31.80 32.00 31.62 31.80 31.98 CUI Nal eee sae booeateaas ASIEN SD sespoane bos Coa DO ETI oe 431.79 31.80 31.99 1 Prepared by nf i Smith (theory = 33.11% As20s). 2 Prepared by . T. Graham. 3 Results by W. B. Pope (U.S. Bur. Chem. Bul. 132, p. 45.) 4 Results by R. C. Roark, (U. S. Bur. Chem. Bul. 137, p. 37. LEAD ARSENATE WITH LEAD ARSENITE. This sample was prepared by mixing known quantities of lead arsenate and lead arsenite. It was designed to test Method III for total arsenic in lead arsenate present as As.O3; only, but through a misunderstanding some of the codperators made this determination on the sample of lead arsenate alone—a sample which contains only a trace, if any, arsenic as As,O3. The results obtained are as follows: 170 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Resulis on lead arsenate with lead arsenite. TOTAL ARSENIC AS As2O5 Rov arivesn jt ENS AsxOs Method | Method | Distillation} °%™* AEE I (a) I (b) method Dean C. Kellog, East Lansing, Mich? s..05225ece segaeene cs oeaceek 30.60 SOLGON | sae cnece 21.35 5.76 30.50 80860) Wiceascn. cs 21.60 5.78 enna ser SG) lteposnpen|) 2alave 5.76 INGE pagebepanacaepeapsaddode 30.55 Gl) y file Sor copoc 21.51 5.77 RAC Roark, Washington, "D: ©o |i eaecee|se see: a 30.48 21.65 7.60 Perea tern| crores 30.50 21.55 7.44 Ag ose SeRepel| spar eievavevsies [lays sists eters 21260) j| craters TAVETARE ce ce cloeiaetcise Sos) e «otal ue atohspesters| factors loPerars 30.49 21.60 7.52 Hugh) Fulmer, Guelph; Canaday| oe rert cel esis eee lee Pees 5.88 BS tan eee as PR or noeracd be 5.83 INE (ha sup OSAP hoSnobaODe +s Food s0au baoso. Decleor op doDEl boo anes sc 5.86 A. J. Flume, Geneva, N. Y 6.35 6.34 6.25 6.34 INV CTA E ei yciatscelsclern tee einse: etal larerserareme etal eetete rahe Sree | echoes he area ice ee Bets 6.32 Geo. E. Holm, St. Paul, Minn....} 30.75 SOS) iscossaces 21.80 7.54 31.10 S3O=50! Wi cerewis-csore 21.76 7.54 INVETA GORE sialic -tacctyos. + ose 30.93 SONG0! 0) Sosvcreen 21.78 7.54 W. H. Rogers and E. R. Tobey, (Opqa) OW Cad anoariae Gerionenedeas laanncodoc DEaooessea smab eco. 21.35 6.11 UNO NP ase Meee ipa o ae Ol 21.27 6.20 IAVOTAPCS He. eocisc celeticae t= + SMe eine «ial ots she Fae ile eats 21.31 6.16 General average.............- 30.74 30.58 30.49 21.55 6.45 Calculated: eaacses eee aoe 30.48 30.48 30.48 21.48 7.75 DISCUSSION OF RESULTS. The results for As,O; are rather discrepant, which may be accounted for in the varying lengths of time the sample was boiled with sulphuric acid. A. J. Flume says in regard to this: “If the samples are boiled less than five minutes with the H,SO,, the results are not reliable.”’ This sample was prepared by mixing 400 grams of the lead arsenate used for the association work this year with 200 grams of lead arsenite. According to the analysis of the referee, this lead arsenite contained: 1917) ROARK: INSECTICIDES lal Per cent As:O3 MobaledrsenictasvAs sa ey ose Ws sa Neh oaks Solee note ae a aeeederocesle ee 23 .63 (Distr l sironymetnod eens ccs ek c.-pni- dace de ee ee eee 23 .66 LAN GENS dBseco 6.6 DOCU tnd TREE Soe ORC ma Ene au ama ha doneere. cio) iNT (Oe O10 Nien cds od oo OOS EOE SEC an ETE cone ar 0.45) 0.39 (Method! Il for lead!arsenate)).\ o<...0.o0..ccsceteeeneuees ORD 7 (Ga ees MOLAIPASO)s Presents) DyAGULETENCEs.)< cia eeiacle vies te siere ets sieceetcierere 23.26 As,0; found directly by Method ITI....................-.---00- 23.15 If we take the composition of the 1915 association sample of lead arsen- ate as 32.00% As.O; with no As,O; present, the sample of lead arsenate with lead arsenite would have the following composition: Total arsenic (calculated as As2O;) = 30.48 2% 32 As2O; only (25652-00025) a (1 X 0.45) =) 21048 23.26 _ As.O; only a6 or 7.75 7.75 As.0; = 9.00 AsO; 21.48 AsO; = 18.69 As.O3 The referee recommends that the method for the determination of As.O3 only be further studied, increasing the time of heating to 20 to 30 minutes and the amount of concentrated sulphuric acid to not less than 5 ec. to 1 gram of sample. The referee also wishes to recommend the distillation method as an official one for the determination of total arsenic in a lead arsenate. It has been found to yield the best results and with the least trouble. An amount of the sample equal to 5 times the amount of arsenic pent- oxid to which 100 ce. of the standard iodin solution are equivalent should be used, and a titration made on one-fifth of the distillate. In this way the number of cubic centimeters of iodin solution used in the titration represents directly the total per cent of arsenic in the sample expressed as arsenic pentoxid (As2Qs). In regard to Methods I (a) and I (b), as well as the present official method (U. S. Bur. Chem. Bul. 107 (rev.), p. 239), while they yield good results on a comparatively pure lead arsenate, they would determine antimony if present, and in the presence of chlorids arsenic present as As.O3 might be lost through volatilization of AsCl;. Most lead arsenates will contain a small amount of arsenite, and there is reason to believe that antimony may be occasionally present. Moreover, in analyzing lead chlor-arsenates, such as those prepared by McDonnell and Smith, where chlorin is a part of the molecule, the danger of loss of arsenic as AsCls, is still greater. The referee, therefore, recommends that these methods be discarded and the distillation method be substituted therefor. In regard to Method IT, while this yields most excellent results on pure samples and even in the presence of As2O3, it is inapplicable in the pres- 172 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 ence of copper and is affected by any substance that will liberate iodin from an acid solution, e.g., ferric chlorid, nitrates, ete. It will also de- termine any antimony present as Sb2O;. As the method is so very quick and simple, the referee recommends that it be further studied, special attention being given to the effect of various impurities that may be present in commercial lead arsenates. A temperature of 35° to 40°, as called for in this method, is not essential, as we have found a temperature of 20° to 25° gives equally as good re- sults. The standardization and determination should, however, be car- ried out at the same temperature. The referee has found that more accurate results are obtained when the thiosulphate solution is standardized directly against a pure lead arsenate or other arsenate, rather than against a standard iodin solution. A pure diplumbie hydrogen arsenate may be prepared as follows: Precipitate diplumbic arsenate by the addition of a solution of lead nitrate to a solution of potassium dihydrogen arsenate (KH:AsO,) which should be in excess. Collect the precipitate by filtration and dissolve in boiling 1 to 4 nitric acid, adding enough of the lead arsenate to secure a completely saturated solution while boiling hot, then filter rapidly through a folded paper filter, and pour the filtrate into 10 to 12 times its volume of cold distilled water. The precipitate should be collected by filtration, dried, powdered, thoroughly dried at 110°, and kept in a glass- stoppered weighing bottle. Prepared in this way thelead arsenate should have the formula PbhHAsO,, in which the theoretical content of arsenic pentoxid (AsO;) is 33.11%. A lead arsenate of known composition is particularly valuable in checking the strength of the thiosulphate solution, but may be used also in the dis- tillation method to check the value of the iodin solution. CALCIUM ARSENATE. TOTAL ARSENIC. METHOD I. Proceed as directed in Journal of the Association of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 11, paragraph 32. METHOD II. Total arsenic present as As20, only. Determine as directed in Method II for lead arsenate, using, however, an amount of the powdered sample just equal to the amount of arsenic pentoxid (As.Os) to which 100 ce. of the thiosulphate solution are equivalent. The results obtained by the chemists codperating in this work are as follows: 1917] ROARK: INSECTICIDES 173 Results on calcium arsenate. TOTAL ARSENIC AS As2O5 ANALYST Method I Method II per cent per cent Hugh L. Fulmer, Guelph, Canada........... 55.50 56.40 55.70 56.40 LGR sesacocscnan0 100 JOC Ra DOP DO MaCn De 55.60 56.40 C. H. Robinson, Ottawa, Canada............. 57.25 57.80 57.30 57.75 57.20 57.70 FAVELA ZO: a. cscs Eirias fo coos SU ae OrEe 57.25 57.75 A: C. Whittier, Newark, Del.................. 56.60 55.60 56.60 55.70 56.80 55.90 LSID aaa Aopr DAC ERR DOCS OO ASb ONS 56.67 55.73 W. L. Latshaw and J. C. Ripperton, Man- 55.60 55.80 hattan, Kans. 55.60 55.50 Rad Soon ecaadee 55.50 Be ons Momo 55.80 PAV CLAD CR Ne ile nisise cierevaizielsmis reins Sastie Facies 55.60 55.65 BeeseeHiume, Geneva, N.)Neees.2 ccs oeccees 56.51 56.50 56.19 56.40 56.59 56.50 fe Gee asters 56.40 EADIGIECUE) oe orp GH tIntG SUBS eae coe Sinan 56.43 56.45 R. C. Roark, Washington, D. C.............. 55.90 55.90 55.85 55.90 Sia ANN anette arelctetae 55.90 LIGIER ON GURU CIG are ESRC OPC D IOC CTIE 55.88 55.90 Dean C. Kellog, East Lansing, Mich......... 56.00 55.60 56.20 55.60 PGES Oust cttr ciohecl-iofeareias ee estas = lessees 56.10 55.60 W. H. Rogers and E. R. Tobey, Orono, Me. 55.32 55.30 DHFS, «a Scleitecses Shore PNET EEO ered <'a/ahcley sie) s sie e Gnisic s\e ie ad dravaerenac 55.41 55.30 Goneraliaverage: 2.7). \.)...: cases se aneeeees 56.22 56.18 The recommendations made in regard to the methods for lead arsenate apply equally to the methods for calcium arsenate. 174 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 ZINC ARSENITE. TOTAL ZINC OxID (ZnO). METHOD I. To 0.5 gram of the powdered sample add 20 ce. of a mixture of equal amounts of hydrobromie acid (specific gravity 1.31) and hydrochloric acid (specific gravity 1.19) and evaporate to dryness. Repeat until all arsenic is removed (two evaporations are usually sufficient), then evaporate to dryness with concentrated hydrochloric acid. Dissolve in 25 ec. of 2N HCl, dilute to 100 cc., and pass in hydrogen sulphid until all lead, antimony, and remaining traces of arsenic are precipitated. Filter, wash precipitate thoroughly with N/2 HCl saturated with hydrogen sulphid, and concentrate the filtrate to small volume by boiling. When the solution is free from hydrogen sulphid, add 1 or 2 ce. of concentrated nitric acid and boil a few minutes to oxidize all Fe™ to Fe™, then precipitate the iron with ammonium hydroxid. Filter, wash precipitate thoroughly with hot water, then redissolve the precipitate in a little hot dilute nitrie acid, collecting in a dry beaker, and reprecipitate the iron as before, filter, wash precipitate thoroughly with hot water, combine both filtrates and washings, add an excess of concentrated nitric acid, and evaporate to dryness on the steam bath to remove ammonium salts. Take up in water, together with a little hydrochloric acid, cool, add 10% sodium carbonate solution drop by drop until the zine solution becomes turbid; then transfer to a porcelain casserole and heat to boiling for a few minutes. Now add a few drops of phenolphthalein solution and sufficient sodium-carbonate solution to produce a distinct pink color, boil for a minute or two longer, then filter from the hot solution. The precipitate should be washed with hot water, first by decantation and then on the filter. The thoroughly washed precipitate on the filter is dried, then carefully ignited (pref- erably in a muffle), and finally heated over a Bunsen burner. From the weight of the ignited zine oxid calculate the per cent of zinc oxid (ZnO) in the sample. (Norr.—This method is not applicable in the presence of calcium.) METHOD II. (Method of Balls and McDonnell.') Dissolve 0.5 gram of the powdered sample in dilute sulphuric acid (1:5), add 50% KOH solution to 20 grams excess KOH, oxidize all the arsenic to AsY with a little sodium peroxid, transfer to a weighed nickel crucible of about 125 ec. capacity, and electrolyze, using a rotating anode. The anode rotation should be about 600 revolutions perminute, and the current 3 to 4 amperes per 100 sq. em. of cathode sur- face. When the zinc is all deposited, which should take 1 to 2 hours, wash the de- posit before interrupting the current by siphoning, then rinse with alcohol and dry to constant weight at 110° in an oven. From the weight of metallic zine calculate the per cent of zine oxid (ZnO) in the sample. Factor: Zn X 1.24476 = ZnO. (Norr.—If the deposit shows a tendency to be spongy, this may be overcome by adding 2 or 3 cc. of a mixture of equal parts of glycerol and ethyl alcohol.) 1 J. Ind. Eng. Chem., 1915, 7: 26-29. 1917] ROARK: INSECTICIDES 175 TOTAL ARSENIC. METHOD I. Total arsenic present as As.0; and As:0;. Proceed as directed in Journal of the Association of Official Agricultural Chem- ists, volume 2 (1916), No. 1, Part II, page 11, paragraph 33. METHOD II. Total arsenic present as As:03 and As0;. Weigh carefully an amount of the powdered sample equal to five times the amount of arsenic trioxid (As203) to which 100 ec. of the standard iodin solution are equiv- alent, transfer to a 250 cc. graduated flask, dissolve in about 100 ce. dilute acetic acid (10 parts glacial acetic acid to 90 parts water), heating to boiling if necessary, then add 4 to 5 grams oxalic acid. Continue the heating for a few minutes, then cool, make to volume, shake thoroughly, filter through a dry filter, pipette 100 ce. of the clear filtrate into an Erlenmeyer flask, add 3 to 4 ce. concentrated sulphuric acid and 1 gram KI, boil down to about 40 ce., cool, remove excess iodin with a few drops of N/20 thiosulphate, then add sodium bicarbonate in excess, and titrate with stand- ard iodin solution in the usual way. The number of cubic centimeters of iodin used in this titration divided by 2 represents directly the total per cent of arsenic in the sample expressed as As2O3. (Note.—In case antimony is present it will be determined and reported as As2Oa according to this method.) METHOD III. Total arsenic present as As20; only. (a) Proceed exactly as directed under (a) Method III for total arsenic in Paris green. (b) Proceed exactly as directed under (b) Method III for total arsenic in Paris green. (In either case, as the content of As.O; in zine arsenite is less than 50%, the iodin must be added from a burette and not from a 50 ce. pipette.) METHOD IY. Total arsenic present as As,0, only. Determine as directed under Method II for total arsenic in lead arsenate. As the amount of As2O; in zinc arsenite is usually small, weigh an amount of the pow- dered sample equal to 5 to 10 times the amount of arsenic oxid (As2O0;) to which 100 ce. of the standard thiosulphate solution are equivalent. If iron salts are present the hy- drochloric-acid solution will be colored, and the use of starch paste is necessary to de- termine the end point in the thiosulphate titration. The standardization of the thiosulphate solution should be made with the aid of starch paste also, if it is used in the determination. (Norr—If any antimony is present as Sb.O; it will be determined and reported as As,0; according to this method.) The results obtained on this sample are as follows: 176 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 Results on zinc arsenite. ZINC ANALYST OxID TOTAL ARSENIC AS As2Oz Method| Method] Method} Method| Method I Il I II TII (a) As:Os Method] Method TI (b)|} IV per cent\per cent 42.00) 0.50 42.93 42.30 42.20 56.38] 42.40 42.93) 0.79 42.20] 0.79 42.10) 0.79 56.72) 41.72 56.72) 41.65 56.72] 41.69 yokes 41.59 41.98 42.60) 41.83 42.48} 41.80 42.48] 41.85 42.58} 41.85 41.83] 0.78 41.85} 0.70 41.85] 0.77 41.75] 0.73 C. H. Robinson, Ottawa, Canada...| 54.64 54.88 AVES Es rani ie cote mie Sota thas eiciotatere 54.76 Hugh L. Fulmer, Guelph, Canada..|...... AVOErage:) cecil ceviin s\sterisisiaell aes A. C. Whittier, Newark, Del........]...... IAVETA RE Sasa e eee chee epic mone W. L. Latshaw and J. C. Ripperton, |...... Manhattan, Sans-y) J) sn | oy ieleoeiee IASVET REE as icrepetelecsievsieieiere ess ako ate fe eral Mepeeee Dean C. Kellog, East Lansing, Mich.| 54.96 54.74 54.36 A VOT ARC Ei ois .ctie heme innate eo 54.69 ACY: Hlame sGenevasnNe ss: 4/7-eeie| cere IA VOLTA BO... iseiecia ciao rare aire ieee ee F. L. Elliott, Washington, D. C..... 56.36 56.52 57.06 57.00 Average cia ts a eee 56.74 J. J. T. Graham, Washington, D. C.}...... AVOLS Ss ie oa cists orerstoensareteretoterocieral ee hears W. J. Morgan, Washington, D. C...]...... Average 177 1917] ROARK: INSECTICIDES Results on zinc arsenite—Continued. ZINC OXID TOTAL ARSENIC AS As2O3 As:0s ANALYST Method} Method! Method III (a) | III (b) IV Method] Method} Method} Method I I I Il per cent|per cent|per cent|per cent|per cent|per cent|per cent E. J. Nealon, Washington, D. C...| 55.48)...... 41.98] 42.18]...... ALENT Easee 552 261F> «3 41.85] 42.31]...... ely Ningasee PUTOLALO woes So cine iate ey ckjelers acta 3 45)-07(Bnoaee 41.92) 42.25)...... AUTO acterse C. H. Walker, Washington, D. C....| 56.35} 56.34) 41.37) 42.48) 41.23) 41.42) 0.72 56.10) 56.49) 41.34) 42.55) 41.23) 41.52) 0.68 PAOLA GE Deitareyss boinc savy en Sates 8s 56.23) 56.42) 41.36) 42.52) 41.23) 41.47] 0.70 R. C. Roark, Washington, D.C..... 5OrS4l yee = 41.80) 42.44) 41.80) 41.80} 0.74 HU Ye omoac 41.69] 42.18] 41.80) 41.80) 0.78 ice | (NORE Nees chs 42.31] 41.70} 41.70) 0.78 Average..... Res BODES iil) (Spee 41.75) 42.31] 41.77] 41.77] 0.77 Were. Rogers and E. R. Tobey, |......|....-- 42.32) 42.59) 41.40) 41.63) 0.70 URE, WE 0 Ot eee (ee (ee 42.45) 42.45] 41.45)...... 0.72 GSR OR Tae OB GUO5 Be SOD DIO ECCS CAE eee 42.39) 42.52) 41.43) 41.63} 0.71 (General average.......-..--.... 55.60) 55.98} 42.01] 42.61) 41.89) 41.98) 0.74 DISCUSSION. Mr. O. B. Winter of the Michigan Agricultural Experiment Station, comments as follows: The methods, with the exception of the two for the determination of zinc oxid, seemed quite satisfactory. In Method II, the anode used was about 5 em. in diam- eter, perforated, and made about 150 revolutions per minute. We used a low current at first, and then increased to 2} to 3 amperes. This was not according to directions and may account for the inconsistent results. In Method I, the solution showed that not all the impurities were removed. This may explain the high results. C. H. Walker, of the Bureau of Chemistry, Washington, D. C., also determined the zinc oxid by titration with potassium ferrocyanid ac- cording to the method of C. Fahlberg (Fresenius Quantitative Analysis 1911, 2: 443-444) and by electrolysis in potassium-hydroxid solution in a nickel crucible after the removal of arsenic, lead, and antimony according to Method I, page 174. By titration he obtained 56.76%, 56.70%, and 56.83%, and by electrolysis, 56.23% and 56.30% ZO. The referee is of the opinion that electrolysis of the zine after the removal of arsenic, lead, and antimony will be found to be a much better procedure than that of Balls and McDonnell. JOUR. ASSOC. OFFICIAL AGRIC. CHEMISTS, VOL. III, NO. 1 178 ASSOCIATION OF OFFICIAL AGRICULTURAL cHEMIsTs [Vol. III, No. 1 This sample contains a small amount of antimony as Sb.0;. The figures in the column headed ‘“‘As.O;, Method IV” multiplied by the factor 1.39353 will give the percentages of Sb.O;, the average being 1.03% The differences in the results for arsenic trioxid (As,O3) by Methods I and II are due to the antimony, which is determined by — Method II, but not by the distillation method, or Method I of the table. The average difference in the results by these two methods is 0.60% As:Oz, which calculated to As,O; gives 0.70%, a figure that agrees very closely with 0.74%, the average of the results by Method IV for As.O; only (really Sb.Os). Results for arsenic trioxid (As,O3) only by the modified methods of C. C. Hedges and C. M. Smith (Methods III (a) and III (b) of the table) agree almost exactly with those for total arsenic by the dis- tillation method. This shows that all the arsenic is present as AseO3 and all the antimony as Sb2O;; furthermore, that the distillation method effectively separates a large amount of arsenic from a small amount of antimony. The only one of the methods for the analysis of zine arsenite which the referee wishes to recommend for official adoption is the distillation method for total arsenic. It is recommended that Method II for total arsenic be discarded and that the methods for zinc oxid be further studied. The other methods have been spoken of under Paris green and lead arsenate. BORDEAUX MIXTURE. (1) Moisture.—Proceed as directed in Journal of the Association of Official Agri- cultural Chemists, volume 2 (1916), No. 1, Part II, page 11, paragraph 36. (2) Carbon diorid.—Proceed as directed in Journal of the Association of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 12, paragraphs 37-38. (3) Copper.—Electrolytic method: Proceed as directed in Journal of the Asso- ciation of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 12, paragraph 389. Thiosulphate method: Proceed as directed in Journal of the Association of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 12, paragraph 40. The results on this sample follow: Results on Bordeaux mixture. COPPER CARBON ANALYST MOISTURE DIOXID = Electrolytic |Thiosulphate per cent per cent per cent per cent A. C. Whittier, Newark, Del............ 4.20 1 PAY Pe ae 12.98 4.24 12520) ica en 12.91 ANGELA GOs noes cco esos RISC Te cites 4.22 WL lie cine eee 12.95 1917] ROARK: INSECTICIDES 179 Results on Bordeaux mixture—Continued. CARBON COPPER ANALYST MOISTURE DIO XID Electrolytic |Thiosulphate per cent per cent per cent per cent R. C. Roark, Washington, D.C....... 4.57 12.68 12.77 12.90 4.72 12.73 12.85 12.83 PAV ETAGE eect <1 ae 8 3 ged ee 4.65 12.71 12.81 12.87 pam aptulatme: | (Gene wale Niece r foraisyaisieise: 2: 2\llssye%s's.c, ccd | sctetesc/alsvers «| Ghoihere o peters 12.87 Pee oot Robo reo an esedeeacon 12.90 BPAaA stats | m epeseresoie: ore areve ete erate 12.87 Nin Goodeeeesd boooeerand poeceeeoe: 12.87 AN IGRICZDs Saab oR SGOD BOC OCOSMCO oOo OSs GGRecEnead Ooncceoead Aaapeaacns 12.88 W. H. Rogers and E. R. Tobey, Orono, 4.84 TBARS) 13.03 13.07 VCMT Bopha FT aio cea eeles 12.69 13.04 13.13 MN er acento ek ew ls vse 4.84 12.62 13.04 13.10 Weneralisve4rage snc cac. 6 tieles sere - 4.51 12.68 12.92 12.94 BORDEAUX MIXTURE WITH PARIS GREEN. (1) Moisture.—Determine as directed for Bordeaux mixture. (2) Carbon dioxid.—Determine as directed for Bordeaux mixture. (3) Copper.—Electrolytic method: Proceed as directed in Journal of the Associa- tion of Official Agricultural Chemists, volume 2 (1916), No.1, Part II, page 13, paragraph 44. Thiosulphate method: Weigh 2 grams of the dry powdered sample, transfer to an Erlenmeyer flask, add 25 cc. of concentrated nitric acid, and heat on the steam bath to disappearance of brown fumes. Dilute somewhat with water and boil for several minutes, then add 10 cc. of bromin water and continue boiling until all bromin is expelled. Neutralize with concentrated ammonium hydroxid and add about 5 cc. in excess. Boil a minute or so and add acetie acid in excess. Cool thoroughly, add about 3 grams of KI (or 10 ce. of KI solution, 30 grams to 100 ce.), and titrate immediately with standard thiosulphate solution in the usual way. Be careful that copper remains in solution. If copper appears to precipitate, it may be redissolved by the addition of a little acetic acid and rubbing the precipitate with a stirring rod fitted with a rubber policeman. Near the end of the titration it is well to add the starch solution in successive small quantities. (4) Arsenic trioxid (As.0;).—Proceed as directed in Journal of the Association of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 13, paragraph 45. C. C. Hedges method, modified: Proceed as directed in Journal of the Associa- tion of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 13, paragraph 46. C. M. Smith method, modified: Proceed as directed above, using dilute sul- phuric acid (1 to 4) instead of dilute hydrochloric. The solution in this case may be heated to boiling. (5) Water-soluble arsenious oxid (As,0;)-—Treat 2 grams with 1,000 cc. distilled water, digesting for 24 hours at a temperature of 32°C., shaking 8 times during the 180 ASSOCIATION OF OFFICIAL AGRICULTURAL cHEMISTS [Vol. III, No. 1 day at intervals of 1 hour. Filter through a dry filter, take a 250 ce. aliquot, make slightly acid with HCl (methyl orange as indicator), then alkaline with excess of sodium bicarbonate, and titrate with N/20 iodin as usual. Make corrections for iodin necessary to produce the same color, using same chemicals and same volume. Calculate all results to original material. The results of the codperators on this sample are as follows: Results on Bordeaux mixture with Paris green. COPPER TOTAL ARSENIC AS As2O3 om pane S WATER- ANALYST > i istil- SOLU- agents Electro- ee Pee Hedges} Smith | Bre lytic method |method| As2Os - per cent|per cent|per cent phate |method per cent|per cent|per cent|per cent|per cent A. C. Whittier, Newark, Del.} 3.56 | 2.59 }...... ads) 0) 30.10} 11.22 AN034|/2200i|e cane 18.15/31 45 leer 30.15] 11.22 3.481262 ||...2.% LEON RSI 5 ieee 30.60) 11.22 AV erage sneer SOOM e202 || ever / eels UGAGURS1250|Fere 30.28} 1.22 Wels batshawland)Jer@sRip=||eeeieriel aerial eles | seer 31530|(31.40- ee See perton, Manhattan, Kans. |..-...}......]......|...... 31220) 31230) >sSeeel eee I SEHEN bap coddde Uses soo taodon||Scoacd|leauoual|laoogos 31. 25)) 31235] eels eee AR Je blume:|Genevan Nevins alecete tral ele bee ehisie rte 17,522) "3095) 2 5.22. seen eee Rien cl tetra taninces 17.429) '32.. 251.6% 1. seein Gene waaletael ener cel eeeeans 17.19)! 32/50) ..55.c0< erased | eee Es eA ere Seualloicachs ieseeener 32/55) |: .0j6,2:a)| Sarees eee INNVEN Se soo napaoeaenspoan esse o|loo0c. cclloaqonc 17.23) 32.31). 3.25.|- nace eee R. C. Roark, Washington, | 3.35 | 2.55 | 18.80] 18.53] 31.93) 31.80] 31.80) 1.89 D. C. BeG5) [PAWS loeo ode 18850| Foes 31.85] 32.00} 1.92 Bsa ob neroeion| Geena smcaea cu 31.90} 31.80} 1.88 INVETAGE alec cites Ts 3.34 | 2.60 | 18.80} 18.52} 31.93} 31.85} 31.87} 1.90 George): Holm) St- Paull) | hi..cille-laaclias stall oe 32.00)...,.. 31.78} 1.81 Ninn SP) pee Udall sete ese cis ecceeioeltasoeaaes 31196| cere 31.80} 1.82 UV Wif 2) 2:14 eee oles irs Te ol oI al (sires es menat aioe c STLO8| 2 Scie 31.79) 1.82 W. H. Rogers and E. R. Tobey,| 2.94 |13.48 |...... 20.15) 31.53}... 5. 32.30) 12.41 OronoyiMe: SA eno SEAS a rer aare 20.33] 31).23].....- 32.50) 12.51 Seer £3 AS. cena |s woe ae|oeue ae ecient Ronee | IAN CTA 0 fa foistcteloions rnyannie veers 2298 Se A OM ee cme 20224) 31441 eee 32.40} 2.46 yeneral average..........| 3.45 | 2.61 | 18.80] 18.20] 31.78] 31.65} 31.57] 1.86 Calleulated).5 arcsec eeepc lemerctere 1SES3|eerier 81.98) 5.6 e050 siesta 1 Omitted from the general average. This sample was prepared by thoroughly mixing 293.5 grams of the 1914 association sample dry Bordeaux with 348.5 grams of the 1914 association sample Paris green. By the distillation method this Paris 1917] | ROARK: INSECTICIDES 181 green was shown to contain 58.91% to 58.93%, average 58.92%, AseOs. The Bordeaux-Paris green should therefore contain 31.98% As:O3. By separating the copper from the arsenic by precipitation with sodium hydroxid and electrolyzing it in nitric-acid solution, the 1914 association sample of Paris green yielded 23.27%, 23.37%, average 23.32%, copper (Cu). Taking the general average of all results on the 1914 association dry Bordeaux for copper by the electrolytic method, which is 13.49%, the calculated amount of copper in the 1915 association Bordeaux- Paris green is 18.83%. Mr A. C. Whittier, of the Delaware Agricultural Experiment Station, makes the following comments: I used approximately an N/20 solution of sodium thiosulphate and got titrations of 108.8, 117.8, and 116.2 ce. I could get no real permanent end point, the blue color very slowly reappearing after each addition of thiosulphate, after 100 cc. had been added. The determination was repeated, using 1 gram instead of 2 grams. Great care was necessary to procure a permanent end point. The thiosulphate was added until no blue color returned after standing one minute. The results obtained by Mr. Whittier, using different amounts of sample (thiosulphate method) are as follows: Per cent copper POST RIMGBATNDICS. -ro. cays aso a/o clever ote zara: aareis\oicye/aveiasiafsietaleis's ce\e.s 6S seve 6 16.77 18.15 17.91 VAVOTA RE oars Gisicn asin «Sissies « o « c,0 012 maiolereleVereveiars orciase meter 17.61 PE SET AITIBB ALIN DLO seater Ntcret si ctere ts oles e aleve ss) ot cCr ee aerate eva eee 18.80 18.86 18.83 LYE oe DOS B BID OO OD AOD Od ODO CeTISD Joop tadhene ou ror 18.83 The results on the 1-gram sample agree closely with the value deter- mined by the electrolytic method by the referee. The referee found that if distilled water which had been recently boiled and cooled was used instead of ordinary distilled water in determining the water-soluble arsenic in this sample, percentages of 2.15, 2.12, and 2.12 As.O3 were obtained. That these results should be higher was somewhat surprising, and emphasize the importance of following the method closely in every detail. Temperature is one of the most important factors in determining water-soluble arsenic and should be closely watched. In or- der that the water may be at 32° at the beginning of the digestion, it should be kept at this temperature for some time before adding the sample to be tested. These methods were studied last year. The referee wishes to recommend that the methods for moisture, carbon dioxid, total arsenic by the dis- tillation method, and water-soluble arsenic be adopted as official and that the other methods be further studied. 182 ASSOCIATION OF OFFICIAL AGRICULTURAL cHEmists [Vol. III, No. 1 BORDEAUX MIXTURE WITH LEAD ARSENATE. (1) Moisture-—Determine as directed for Bordeaux mixture. (2) Carbon dioxid.—Determine as directed for Bordeaux mixture. (83) Copper.—Proceed as directed in Journal of the Association of Official Agricul- tural Chemists, volume 2 (1916), No. 1, Part II, page 13, paragraph 44. In determining copper electrolytically in a Bordeaux-lead arsenate or a Bordeaux- Paris green, be sure that all the arsenic is in the ‘‘ic’’ (As”) form. (4) Lead oxid.—Proceed as directed in Journal of the Association of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 14, paragraph 52. (5) Total arsenic pentoxid (As20O;).—Proceed as directedin Journal of the Associa- tion of Official Agricultural Chemists, volume 2 (1916), No.1, Part II, page 14, paragraph 53. (6) Water-soluble arsenic oxid (As2.0;).—Proceed as directed in Journal of the Association of Official Agricultural Chemists, volume 2 (1916), No. 1, Part II, page 11, paragraph 31. The following results were obtained on the sample: Results on Bordeaux mixture with lead arsenate. TOTAL WATER- ANALYST MOISTURE quneon COPPER pas rere ae oe per cent per cent per cent per cent per cent per cent A. C. Whittier, Newark, Del.. 2.36 CDOS N(- aeaeeen| peoeeiar aes 14.10 0.20 2.44 Chea) Ptarsteacin ein oF 14.10 | 0.17 a sca etal Sa eeTaeelaie HO erckenceetetel | ee ele reiare 14.15 | 0.20 AMET ARE nce cole ae ciecrs ciel eitelctiietetistebtetetarelels 20.98 21.09 JAVOL ARES cs cacti erscen cia re ie toree re alese we athelecdekencr elle tatelajsneteraseralotof=tatete 21.04 HAs Meyers) Ss usttalos ting vir ctctereratsiesoletetersPataxe|| erarsroyetatetel et siete Folens 420.40 420.53 20.71 20.81 PAV YC RGR AA Ona eHe a Oe MOOD AOU odao | nda Cooma ac o10D00Re 20.61 General averages... 2 cei. ceactenisircniostes : 20.83 12 grams distilled; 0.2 gram analyzed. 210 grams distilled; 0.5 gram analyzed. 210 grams distilled; 0.25 gram analyzed. 3 F A Due to the fact that the precipitate is so profuse, it is believed that there might have been a slight loss in filtering and a low result obtained. 1917] ROARK: INSECTICIDES 185 RECOMMENDATIONS. It is recommended: (1) That Method I for total arsenious oxid in Paris green (U. 8. Bur. Chem. Bul. 107 (rev.), pp. 25-26), and as modified on page 157, be discarded. (2) That the distillation method for total arsenic as described on page 158 be adopted as official and designated Method I. (3) That the modified methods of C. C. Hedges and C. M. Smith for the determination of total arsenic as AsO; only in Paris green (Methods III (a) and III (b), p. 158) be not adopted as official, but that they be used as nonofficial methods for the quick determination of the approxi- mate amount of As,O3 present. (4) That the present official method for the determination of total arsenic in lead arsenate (U. S. Bur. Chem. Bul. 107 (rev.), p. 239) and as modified (Methods I (a) and I (b), p. 163) be discarded. (5) That the distillation method as described on page 171 be made official for the determination of total arsenic in lead arsenate. (6) That Method II for the determination of total arsenic as AsO; only in lead arsenate, pages 163-164, be further studied. (7) That Method III, page 164, for the determination of As.O3 only in lead arsenate be changed so as to require a boiling of from 20 to 30 minutes with 5 ce. of concentrated sulphuric acid to each gram of sample, and as so modified be further studied. (8) That the distillation method as described on page 172 be adopted as an official method for the determination of total arsenic in calcium arsenate. (9) That recommendation (6) of this report apply equally to calcium arsenate. (10) That the distillation method as described on page 175 be adopted as official for the determination of total arsenic in zinc arsenite. (11) That Method II, page 175, for the determination of total arsenic in zine arsenite be discarded. (12) That recommendation (3) apply equally to zinc arsenite. (13) That recommendation (6) apply equally to zinc arsenite. (14) That Methods I and II for the determination of zine oxid in zine arsenite be further studied. (15) That method (b) for the determination of moisture in Bordeaux mixture, Bordeaux-Paris green, and Bordeaux-lead arsenate mixtures, when in the form of pastes, as described on page 178 be adopted as official. (16) That the method for the determination of carbon dioxid in Bor- deaux mixture, Bordeaux-Paris green, and Bordeaux-lead arsenate, as described on page 178 be adopted as official. 186 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 1 (17) That the electrolytic method for the determination of copper in Bordeaux mixture as described on page 178 be adopted as an official method. (18) That the thiosulphate titration method for the determination of copper in Bordeaux mixture as described on page 178 be adopted as an official method. (19) That the method for water-soluble arsenious oxid in Bordeaux- Paris green as described on pages 179-180 be changed so as to require carbon-dioxid-free water and as so changed be adopted as a tenta- tive method. (20) That the distillation method for the determination of total arsenic in Bordeaux-Paris green as described on page 179 be adopted as an official method. (21) That recommendation (3) of this report apply equally to Bor- deaux-Paris green. (22) That the electrolytic method for the determination of copper in Bordeaux-lead arsenate as described on page 182 be studied further with reference to its applicability to the determination of copper in both Bordeaux-Paris green and Bordeaux-lead arsenate, particular attention to be given to the effect of the various impurities which may be present in commercial samples. (23) That the thiosulphate titration method for the determination of copper in Bordeaux-Paris green as described on page 178 be further studied. (24) That the method for water-soluble arsenic oxid in Bordeaux- lead arsenate as described on page 182 be adopted as a tentative method. (25) That the method for the determination of lead oxid in Bordeaux- lead arsenate as described on page 182 be further studied. (26) That the silicotungstic-acid method for the determination of nicotin as described on page 183 be adopted as an official method. (27) That the cooperative work on insecticides for next year include a study of the following: (a) A method other than an electrolytic one for the separation and determination of copper and lead in a Bordeaux-lead arsenate mixture. (b) Methods for the determination of the principal ingredients in zine- arsenic compounds, alone and in combination with Bordeaux mixture. The association adjourned at 12.30 p. m., to reassemble at 1.30 p.m. PROCEEDINGS OF THE THIRTY-SECOND ANNUAL CONVENTION OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS, 1915. THIRD DAY. WEDNESDAY—AFTERNOON SESSION. REPORT OF COMMITTEE ON NOMINATIONS. By W. B. Exxert (Agricultural Experiment Station, Blacksburg, Va.), Chairman. The committee submitted the following nominations for officers for the year ending November, 1916: president, R. N. Brackett, of South Carolina; vice-president, J. K. Haywood, of Washington, D. C.; secre- tary-treasurer, C. L. Alsberg, of Washington, D. C.; additional mem- bers of the executive committee, W. J. Jones, jr., of Indiana, and E. B. Holland, of Massachusetts. The secretary was instructed to cast the unanimous ballot of the asso- ciation for these officers. REPORT OF COMMITTEE ON RESOLUTIONS. By R. J. Davinson (Polytechnic Institute, Blacksburg, Va.), Chairman. Resolved, That the cordial thanks of this association be extended to our secretary, C. L. Alsberg, the executive committee, and the publishers, the Williams & Wilkins Company, of Baltimore, for the culmination of their efforts which has resulted in the successful publication of the Journal of this association. Resolved, That the hearty thanks of this association be extended to the committee on editing methods of analysis for the painstaking care with which this laborious and extensive work has been conducted and presented for publication. Resolved, That the thanks of the association be extended to the Chemical Society of Washington for the entertainment on Tuesday, November 16, 1915. Resolved, That this association extend cordial thanks to W. D. Bigelow and the National Canners Association for their hospitality. Resolved, That this association express to the secretary’s assistant, Miss N. A. Parkin- son, its sincere appreciation for her valued assistance extended so charmingly to all of its members. 187 188 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 2 Resolved, That the thanks of this association be extended to the Raleigh Hotel for the use of the banquet hall and other conveniences and for the courtesies shown to the members of this association. Resolved, That the association hereby expresses to President, Jones its appreciation for his unfailing courtesy and impartiality in conducting the difficult affairs of his office. The report of the committee was approved. No report was made by the associate referee on nitrogenous compounds of soils. REPORT OF COMMITTEE OF REVIEW ON THE ANALYSIS OF LIME SULPHUR SOLUTIONS. By R. J. Davipson (Polytechnic Institute, Blacksburg, Va.), Chairman. [t is recommended— (1) That the present referee be requested to furnish the chairman of the committee on editing methods of analysis with some methods to be printed by that committee. (2) That it would be wise to discontinue this committee and refer the work to the regular referee on this subject. The recommendations were approved. No report was made by the referee on medicinal plants and drugs. No report was made by the associate referees on synthetic products, medicated soft drinks, and medicinal plants. REPORT ON ALKALOIDS. By H. C. Funuer (Institute of Industrial Research, Washington, D. C.), Associate Referee. The general plan of study adopted involves the determination of cer- tain of the alkaloids which are commonly found in preparations in general use among physicians and which are made in a very large way by the manufacturing pharmacist. It was decided that strychnin should be the first alkaloid considered. Three samples were examined: (1) A carefully prepared mixture of strychnin sulphate and milk sugar; (2) a sample of carefully prepared tablets containing 0.1 grain of strychnin sulphate; and (3) a sample of carefully made tablets containing 0.01 grain of strychnin sulphate. Two methods were used. 1919) FULLER: REPORT ON ALKALOIDS 189 INSTRUCTIONS TO COLLABORATORS. Employ both Methods I and II for determining the strychnin sulphate in Samples 1, 2, and 3. Take 0.3000 gram of Sample 1. Take 10 tablets of Sample 2, weighing carefully. Take 25 tablets of Sample 3, weighing carefully. Report results as follows: Percentage of strychnin and strychnin sulphate in all three samples; grains of strychnin sulphate per tablet in Samples 2 and 3. Method I. Transfer a carefully weighed amount, 0.3000 gram, of the powder to a 200 ce. Squibb separator and moisten with 5 cc. of water. Add 1 cc. of stronger ammonia water. Agitate with 25 cc. of chloroform and allow to stand until separation is complete. Draw off the chloroform into a second separator and repeat the agitation twice with 25 cc. portions of the solvent. After combining all of the fractions, wash the combined chloroformic solutions by agitation with 10 cc. of water and allow to stand 15 minutes. Introduce a pledget of absorbent cotton into the stem of the separator and run off the chloroform into a tared dish, but do not allow the wash water to enter the orifice of the stop-cock. Add 10 cc. of chloroform, and when the water has entirely risen to the surface, run off the chloroform into the tared beaker. Wash off the outer surface of the stem of the separator with a little chloroform and then evaporate over a steam water bath, using a fan or blower and removing from the bath as the last portions evaporate to avoid decrepitation. Dry at 100°C. to a constant weight and weigh as strychnin. Check the weight of strychnin by dissolving the residue in neutral alcohol, adding an excess of N/10 sulphuric acid and titrating back with N/50 potassium hydroxid. Strychnin to strychnin sulphate 1.2814, according to U.S.P. One cubic centimeter of N/10 sulphuric acid is equivalent to 0.0334 gram of strych- nin and 0.0428 gram of strychnin sulphate. Method II. Yollow the procedure of Method I down to, but not including, the washing of the combined chloroform extract with water. Discard the alkaline solution remaining in the first separator. Treat the combined chloroform extracts with 10 cc. of N/1 sul- phuric acid and agitate. Allow to stand until separation is complete and collect the chloroform in a second separator. Repeat the extraction with N/1 sulphuric acid twice more, discard the chloroform, and combine the acid fractions. Add stronger ammonia in excess, cool, and shake out with three successive portions of 25 ec. each of chloroform, finally combining all fractions. Wash the combined chloroform solu- tions by agitation with 10 cc. of water and allow to stand 15 minutes. Draw off the solvent through a pledget of absorbent cotton into a tared dish and finish the determina- tion precisely as described in Method I. 190 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. ITI, No. 2 TABLE 1. Cooperative results on Sample 1. (Containing 17.76 per cent of strychnin sulphate and 82.24 per cent of milk sugar.) STRYCHNIN STRYCHNIN STRYCHNIN STRYCHNIN ANALYST ALKALOID SULPHATE ALKALOID SULPHATE BY WEIGHT BY WEIGHT VOLUMETRIC | VOLUMETRIC METHOD I. B. H. St. John, Wm. B. Warner Co., per cent per cent per cent per cent Ste LouissiMo; were. ee eee 14.10 18.07 12.93 16.57 C. K. Glycart, U. S. Food and Drug : Inspection Station, Transportation ead Hea HE re Building, Chicago, Ill............... 5 " 3 F. W. Heyl, Upton Chemical Co., Kalamazoo;-Muchsas--6 on. 14.93 19.13 17.12 G. H. Briggs, Institute of Industrial 14.53 18.62 13.50 17.30 Research, Washington, D. G........ 14.50 18.58 13.72 17.58 J. B. Luther, U.S. Food and Drug In- spection Station, U. S. Appraiser’s Stores, New York, N. Y............ 14.33 18:36" /8) “ 2 Se a H. B. Mead, U.S. Food and Drug In- spection Station, U. S. Appraiser’s Stores INewmVOrkei Nec. inten 14.07 18.04 13.50 17.1 EK. M. Bailey, Agricultural Experiment Station, New Haven, Conn......... 14.45 18.52 14.00 17.92 H. C. Fuller, Institute of Industrial Research, Washington, D. C........ 14.51 18.59 A Verage cn: eaihh< acetic et. eer 14.36 18.40 12.91 16.59 METHOD I. BoBSStoJohn 27 Ae a eee 14.57 18.66 13.60 17.42 Galk:Glyeart!.5. ctioonive stiches Gas 14.26 18.26 12.54 16.07 13.77 17.64 «oth: 9 | er BS Wie Heyli.. nee a Oe eee 15.1 LOS" Bee 17.26 GB! Briggs: a2 e ee oo. eee 14.50 18.58 13.72 17.58 14.40 18°45. |) Seen sl Ji Bainuthers 1, Saab teas cata oe 14.50 18258. 90]\) "eee HiBoiWeadt':.:. tee ire ones Seite 14.40 18.45 13.0 16.66 EB: MiepBailey ii./5teaveeie sey iS tose 14.43 18.49 14.03 17.95 HaiGs Pullers: sae see oe eee 13.83 W772 ah Reo Ue AVCLARC hoi nic ccf ee 14.37 18.39 13.37 17.14 1919) FULLER: REPORT ON ALKALOIDS TABLE 2. Cooperative results on Sample 2. (Strychnin sulphate—1/10 grain tablet.) swans | emcee yee ar yar a, muecom |p Serene METHOD I grain j grain per cent per cent per tablet per cent per ceni per tablet BoH:St. John || 7.54 9. 0.1004 7.08 9.0: 0.093 : ls 7.58 9.71 0.099 6.47 8.28 0.085 C.K. Glycart..) 73) 9.37 0.096 6.50 8.32 0.087 8.34 10.50 0.096 me ou Ae F.W.Heyl....| 826 5 a RI lie «7 933- eee. ; 8.19 co ae ae 9.48 F 7.77 9.95 0.1028 7.34 9.40 0.097 C. H. Briggs...) 7'97 10.09 0.1048 7:30 9.35 0.097 J.B. Luther ...} 7.55 9.67 | 0.0978 | | HeB-Mead....| 7.54 9.66 | 7.07 9.06 | E.M.Bailey...| 7.81 10.01 0.103 | 7.59 HUG: Puller-..| 7.64. 9.78 0.103 7.22 9.25 0.097 Mverage..... 7.78 9.85 0.1003 | 7.07 9.13 0.092 METHOD Il. B.H.St.John.| 7.45 9.55 0.1000 7.04 9.03 0.0942 ; | 7.24 9.27 0.096 6.49 8.31 0.086 C. K. Glycart. | 691 8.84 0.092 6.49 8.31 0.087 F.W.Heyl....| 8.41 10.70 O:ag2 |e re aes (ee 7.77 9.56 0.1028 7.34 9.40 | 0.09716 C. H. Briggs..-| 7'93 10.16 0.1058 7:30 9.35 0.09716 Gee) Luther...| 7:76 9.94 oo7ew less) aol) 5. Pe Mipad....| . 7255 67" ot 7.10 9.10 E.M.Bailey...| 7.67 9.83 0.101 7.23 H.C. Fuller...| 7.61 9.75 0.099 6.23 7.99 0.082 Average..... 7.63 9.72 0.098 6.99 8.95 0.090 192 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 TABLE 3. Cooperative resulis on Sample 2. (Strychnin sulphate—1/100 grain tablet.) tenner © | aaraconn, | of PTEMBER (| -TAeaconpan| «ATOR Oe METHOD I. B. H. St. John . Pogs Pr 08 verte OTD 70.03, oer C.K Glycarts-\. Oe) |) too ite. |. 068 t 0.80/H eee EW. Heyl Eieeg lh Wren eee C.H, Briggs...) t-(Gig4 vil! aos eal ule yy /o%e <1 - Oey Tene J. B: Luther... 0.87 Pet 0.011 H. B. Mead.... 0.91 ilbily/ 0.62 0.81 E. M. Bailey... 0.88 1.18 0.011 0.86: || «:h., Seo H. C. Fuller.... 0.84 1.07 0.0106: i ack i) ec eee Average..... 0.87 Laz 0.011 0.71 0.91 0.0093 METHOD Il. B. H. St. John . 0.86 1.10 0.0110 0.68 0.87 0.0095 C.K, Glyeart..|° G75 |. Ope olin. oly oe2 ele | ogo) lame F. W. Heyl... . 1.07 1.37 0.013 0.98 0.0098 CH. Briggs... ne is 0.011 om on 0.0099 J. B. Luther. . . 0.82 1.07 0010 °° |i ase | See H. B. Mead ... 0.85 1.09 0.61 O71 0 eee E. M. Bailey... 0.84 1.08 0.011 082 | ee. Ti eee H. C. Fuller... . 0.82 1.06 0.010 0.86 1.09 0.010 Average..... 0.85 1.09 0.012 0.73 0.91 0.0094 1919} BUC: TEST FOR STRYCHNIN 193 CONCLUSIONS. It is evident from a study of the above figures that the determination of strychnin by titration can not be recommended. Average results indicate that there is no advantage in using the second method, which considerably lengthens the determination. Individual results, however, indicate that there is some advantage in using Method II. It is recommended that in conducting assays for strychnin, reliance be placed on a gravimetric determination and not on a determination obtained by volumetric means. It is recommended, further, that another year be devoted to the study of methods for determining strychnin in tablets, with a view to incorporating further details which may improve the description of the process in such a way that individuals will be able to obtain more concordant results; and, furthermore, that the study of the determination of strychnin be extended to more complex mixtures. DELICATE TEST FOR STRYCHNIN'. By H. E. Buc (Bureau of Chemistry, Washington, D. C.). PREPARATION OF ZINC. Treat granular zinc with a little concentrated hydrochloric acid so as to clear the surface. Pour off the acid. Cover the metal with 1% tartar emetic, shaking occasion- ally during 1 hour. Add a saturated solution of mercuric chlorid (1 ec. for every gram of zinc). Add a few drops of concentrated hydrochloric acid. After 30 minutes pour off the solution and wash thoroughly; dry. TEST. To the dry extracted alkaloid, or any of its salts, or to the aqueous solution, the volume of which is to be about 0.5 cc. or less, add 0.5-1 gram of the zinc amalgam and 0.5 cc. of concentrated hydrochloric acid. If the amount of strychnin is very small (less than 0.01 mg.), allow to stand 15 or 20 minutes. With larger quantities much less time is required. (To save time, after about 2 minutes test a few drops and, if negative, allow to stand full time.) Pour off the solution from the zinc, taking care not to carry along any zinc particles. Add by drops a 0.02% solution of potassium ferricyanid (K;FeCy,). A pink to rose red coloration indicates strychnin. Large amounts of some alkaloids and other organic substances interfere by reacting with ferricyanid (K;Fes(CN)«). In the absence of interfering substances this test will indicate about 0.001 mg. of strychnin. H. E. Bue (Bureau of Chemistry, Washington, D. C.), submitted a paper on “The Estimation of Strychnin in Presence of Quinin.”’ 1 Modification of Malaquin’s test. 194 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 PRELIMINARY STUDY OF SOME OF THE PHYSICAL AND CHEMICAL CONSTANTS OF BALSAM PERU. By E. C. Merriii! (U. 8. Bureau of Chemistry, Washington, D. C.), Associale Referee on Balsams and Gum Resins. A preliminary study of the method for the determination of the iodin value of the cinnamein of Peru balsam was conducted. Some physical constants, such as viscosity, surface tension, optical rotation, and refrac- tometer readings, were also studied superficially. IODIN VALUE OF CINNAMEIN?. Obtain cinnamein from Peru balsam by shaking out the ethereal solution of the balsam with 2 or 3 successive portions of 5% potassium hydroxid in the same manner as in determining the unsaponifiable material in a soap. Wash the ethereal solution once or twice with 10 mils of water and filter through cotton, drive off the ether, and dry the cinnamein in vacuum desiccator over sulphuric acid for about 12 hours. Take about 1 gram of the dried cinnamein for the iodin value, weigh into a glass capsule, transfer to an iodin number bottle, add 10 mils of chloroform and 30 mils of Hanus solution. After standing 30 minutes add 100 mils of water, followed by 10 mils of 15% potassium iodid solution, and excess of iodin titrated with thiosulphate in the usual manner, using starch indicator when near the end point. Carry out a control, using the above stated amounts of reagents and same time for reactions. Results on artificial and true Peru balsam. PERCENTAGE SAMPLE DESCRIPTION Gee ame ENCES SSE (HANUS) (wigs) wean A Labeled synthetic Peru balsam........... 39.47 49.34 25 B Presumablyttrue'y.(.... oseiere eters eerie ei 22.07 42.64 73 Cc Presumably true). . =... (yen. cee ats 26.53 39.25 32 D Presumably trues of \.as piers ese mare 25.96 38.34 49 E Reported as synthetic..................-. 4.70 3.9 oye F Reported as adulterated................. 21.53 31.25 45 G Known to be artificial*.................. 26.51 32.80 23 *Sample from W. O. Emery. Obtained from a German manufacturer. The same quantity of Wijs’ solution and other reagents was used in the Wijs as in the Hanus method. The results on these two methods suggest the need of further investigation of the Hanus method before submitting for collaboration. INFLUENCE OF AGE OR EXPOSURE TO AIR ON IODIN VALUE. The following table indicates the influence of age or exposure to air on the iodin number of cinnamein in balsam Peru. Column 3 shows the ‘ Present address, United Drug Company, Boston, Mass. 2 Method as at present employed. 7) 1919) MERRILL: BALSAM PERU 195 influence of treating the balsam with a current of air for 24 hours and Column 4 of shaking the balsam with hydrogen peroxid for 24 hours previous to extraction of cinnamein. Results when subjected to oxidizing influences. IODIN IODIN NUMBER (WIJS) IODIN NUMBER (wIJs) SAMPLE NUMBER AFTER TREATMENT WITH AFTER TREATMENT WITH (wigs) AIR CURRENT HYDROGEN PEROXID in Dodds See ee et 49.34 44.67 45.13 Biseio bate RCCL RE Sees 42.64 32.63 32.64 (Cie sos BE a Sone eee 39.25 36.28 38.50 Dea oo See eS So cee ae “38.34 36.67 35.51 LS eben: 5, GE ae AES Oe Se 3.90 1.60 3.20 INFLUENCE OF TIME ON IODIN VALUE. The time factor for absorption of iodin by Hanus solution has been investigated. Results indicate that the time at present employed (30 minutes) is insufficient for complete absorption. Results showing influence of time on absorption of iodin. IODIN TINE | VALUE | dohminutes eee yes en? i). ae | 17.70 30 minutes.......... > soe Ee eee: GUT eee ey ae a Baus ACEO | 23.34 ZiNOUTSI Er 9> 5. 200s ST | 23.80 SINOUTS EEF. oiel. ne Sige oa tA ae 27.36 ELING)iTES noo cate Meee MERE han | | 27.00 VISCOSITY OF PERU BALSAM. No marked difference in viscosity between true and artificial Peru balsam was noted. The following table gives a measure of relative viscosity on the samples studied. Comparative table showing time of efflux. SAMPLE TIME | seconds Lie Mee een ei Re es Re nee 31.4 B Fes a Ee NDB alata Sake apy act. 90.0 (Sem OCEANS Gee eee | 154.2 D AS RO Sa Cee ions 5 ea | 94.6 B SSE nga es hs BIN ea chic aro Naty CP aE CS | 88.0 BES oy Bess et Ney AATEC ay Bee 112.0 ree GER he i ig ca | 28.0 * Artificial. 196 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 2 SURFACE TENSION OF BALSAM PERU. The relative surface tension of the above samples was estimated by means of the ordinary stalagmometer. Since the surface tension is pro- portional to the drop weight, results are reported as drop weights in milligrams compared with water. Resulls showing surface lension values. SAMPLE weDROP | WATERED mq. ASS 5 bia RW x Wale ete peer aoc put eer eal: Bis Meaene. aie 35.6 0.283 1 SR, Ine eee a ORR DA ee rte en APM, Sn es 42.6 0.338 GEES IES REA ORES ROA a ea I eae LS AI 41.3 0.329 | DR OEE Gon ihcr h Aac RREPE hts cee GOES renin antag 42.7 0.339 | Di eee eR te Si ee SO mE OY To 3 Ae ae coro Te S 44.6 0.354 Bees ics trace hynls ss oe os OTR Cee Ce ECE Gon aR oe 48.3 0.383 CS a ie i 1 ee RG SS Be Ae alban RU ER Od ee tN 39.9 0.317 Wistter atr25RA5 00) Re EES, RRR eo. Pee eee 125.8 1.000 @ Artificial. There appears to be a general digression in the values for genuine as compared with artificial balsams. While the results obtained are not conclusive on account of the limited number of samples studied, they indicate the possibility of the surface tension value being useful as a factor in connection with the other constants of Peru balsams. OPTICAL ROTATION OF CINNAMEIN IN SOLUTION. The optical rotation of solution of cinnamein in benzol was determined. Solutions containing 2 cc. of cinnamein in 10 cc. of benzol were polar- ized in 100 mm. tubes. Polarization in angular degrees at 25°C. SAMPLE ROTATION OR oe, ony nse ae + 0.35 Bae tect Sei eee roe aio meracutbor | 0 (Ora iia is Ra Jen NR cE re Mt) | + 0.10 De SSR eRe Aaa chy SIRE ts Pe | 0 BGS ; cosh nioue poucortah toe eee + 0.35 | ic eee Pe ree Ota a rs + 0.17 LO os atte Sa nats ae ne ICE ae No reading ® Artificial. The cinnamein obtained from the artificial product appears to give a slight dextrorotation, which is also noted to a lesser extent in one of the samples reported as genuine. The cinnamein of pure Peru balsam shows little or no rotation. A ~I 1919| MERRILL: BALSAM PERU 19 REFRACTOMETER READINGS OF CINNAMEIN. In the case of the artificial Peru balsams, the refractive index of the cinnamein is lower than 1.57; in the natural Perus the figure is above 1.57. This factor may prove useful in the final consideration of the constants of Peru balsam. Refractometer readings. nares aT 25°C. Ao arctan Re IERIE ERC pare TA 1.5450 Beetle tha Pea esate ot sreitve tpech. ae ae 1.5765 (Gide 3 aeons foe Ae Bete 1.5710 iD eee rete Saeeee Ne Eee 1.5720 Bia persed. ..2 hs An ce: Nhwaweke OLOG20 Ph -piae ae I oe ee, 10.9061 COST UTUTTES N |s ae sae eee oor O0G215 a as ae INVECARO Se |W Mana ek eh des 0.0624 0.0627 100.48 Set C: INOS 1S .24 5.5 9.2229 0.0418 ON 263%|| Steere Mh ees eee e|| | et INO 24 or Rt 11.1070 0.0509 OM 2s asepraaely Gated See Wet: ING LBs oy: 10.8454 0.0499 0.1282 Dae tok Po Rereoman | Meee oc Average t):|! PILE OMNES OM27S. PE een EO Bt ln Eee weet INO. AG)... 10.6190 (0 a) eee aeons se 0:06285 | eee INO 1 7 ss). 11.5529 0.1141 owl yeeocedotc C0613 | te see. IND ATS Ah He sic 11.0007 OWT 26)0i|) Aes FAR OMe Zz ge | Ons | Hae | Soe] Boe & | zha|zZ g Bee se eiSien | omersis 3 Zane < =| az Bee | Fae) gaz | baz = i} CC) < = Zz < fe = SAMPLE 1 DRIED BLOOD per cent| per cent| per cent| per cent| per cent| per cent per cent ne B: Deemer. .....2.. 15.07 | 12.84 | 12.48 | 0.217} 12.12} 9.16] 76 Ber ae C. G. Remsburg...... . sate (PLZT 2 D272) | easel ae Ee af ep He Gab Inman: |... ... .)...; 14.97 | 12.78 | 12.75 | 0.240} 12.23) 9.42] 77 0.99} 92 O38 Jensen’. . 25. ..4: 15.97 | 12.46 | 12.27 | 0.210} 12.06; 9.26) 77 1.20} 90 V.B. Hausknecht..... 11.85 | 12.88 | 12.80} 0.200 | 12.56) 9.50| 76 1.30} 90 H. A. Hudgins........ OY) ER PAG) eee. lay 15 ai 1.30) 89 G. F. Anderson........| 15.85 | 12.76 | 12.42| 0.21 | 12.32] 9.24] 75 0.78} 93 ReaD" Caldwelliac.. 3.15 14.36 | 12.62 | 12.59} 0.10 | 12.02] 8.61 72 1.16} 90 ED: Spears: . 13... ./.28 15.01 | 12.76 | 12.80| 0.21 | 12.08) 6.58) 54 1.60] 87 Webi Ehrom . 2k. Ax: 15.74 | 12.22 | 12.62} 0.23 | 11.69] 8.31) 71 3.38 he eS Chilton’ . 25... 4.2 14.07| .... | 12.38 | 0.34 | 11.93} 9.44) 79 1.37] 89 EW. Bradley ..\.....:20- 12.35 | 12.88 | 12.78 | 0.22 | 12.32] 7.91) 64 0.80} 93 R. E. Ingham......... 15.06 | 12.93 | 12.73 | 0.17 | 12.20) 5.31 44 1.60| 87 E. E. Sawyer..........} 14.01 | 12.58 | 12.26] 0.20 | 11.92] 8.83] 74 0.87} 93 SAMPLE 2 TARTAR POMACE AND DRIED BLOOD ReB: Deemer:.....-.. 10.57} 5.58} 5.38 | 0.54 4.68) 2.28] 49 Pee C. G. Remsburg....... Seah DAG 544 Bs Bee Laie Gam. Inman. . 2... 2. 10.38} 5.36) 5.43} 0.49 4.87] 1.79| 37 1.51 69 Os Jensen. 2: <2 12.27| 5.19} 5.22) 0.52 4.69} 2.48) 53 1.52| 67 V. B. Hausknecht..... 8.20} 5.50} 5.48 | 0.53 4.97) 2.32] 47 1.82) 63 A. Hudgins... .......: 7.62 Sie. ||| “OE 4.84 aoe 0.96| 60 G. F. Anderson........ 11.15| 5.34] 5.26} 0.49 4.77| 2.48| 52 1.37) 71 R. D. Caldwell........ 11.25] 5.38] 5.41 | 0.42 4.91] 2.27) 46 1.45} 70 HED. Spears: . 25. . 2... 10.30) 5.41) 5.32) 0.49 4.66| 1.26] 27 0.65} 86 Weep Ebron:.).: 2 2.. 11.72| 5.29] 5.41 | 0.52 4.51} 1.56| 35 2.05| 54 HES:. Ghilton'. . 2)... 10.61 ae 5.10 | 0.58 4.66} 1.92) 41 1.80} 61 PeaaW.. Bradley... 2.0. 8.33} 5.54] 5.50) 0.51 5.04} 1.93) 38 1.26) 75 R. E. Ingham......... 10.41| 5.48} 5.49 | 0.42 5.01} 1.84| 37 1.83] 63 E. E. Sawyer.......... 9.79| 5.40} 5.24! 0.50 4.71} 2.09) 44 1.48} 68 SAMPLE 3 NITROLENE R. B. Deemer......... 6.50} 7.26| 7.02} 0.22 2.60} 1.09} 42 ere he C. G. Remsburg....... 7.15} 7.10} 0.21 PAA ae hogs Be Coeheinman. ........ 6.30} 7.10) 7.10} 0.24 2.82| 0.91} 32 0.64) 77 ORE Jensen’... 2. 5.5 =. 7.46| 6.91] 6.87} 0.21 2.49; 0.90} 36 0.74| 70 V. B. Hausknecht..... 6.15| 7.07} 7.13 | 0.21 2.51] 0.91) 36 0.70| 72 HEA. Hudgins. 7... ... GSSille ae EAS) See 2.84] ... 0.51) 82 G. F. Anderson........| 7.14] 7.02] 7.08) 0.21 2.63| 1.03) 39 0.55) 79 R. D. Caldwell........ 8.58 | 6.94} 7.04} 0.23 2.29| 1.02} 44 0.61| 734 HD: Spears... 2... - 6.75 | 7.07| 7.00} 0.17 2.46] 1.31) 53 0.46) 81 Web) bron). 2.0.20). 7.78} 6.86. 7.03 | 0.21 2.44| 0.59|) 24 0.65} 73 es. Chilton. o..0 «.2.- - 1.30 15.30 15.28 Drepesawyersis cr. Sees ak 8. eee 0.93 15.90 15.80 DISCUSSION OF RESULTS. JONES AND STREET METHODS FOR NITROGEN ACTIVITY. TABLE 1.—Samp ces 1-5, INCLUSIVE. In the discussion of these results only the total nitrogen in the column headed “‘Gunning-Arnold method with copper’’ will be considered in this connection. Water-insoluble organic nitrogen—On the whole, the results are fairly satisfactory. They appear more uniform than those obtained last year on the same samples. Perhaps this is due in part to the use of the Gunning-copper instead of the Gunning-mercury method this year, as it appears from comparative work on total nitrogen that there was less variation with the former than with the latter method. Nitrogen liberated by alkaline permanganate (Jones method).—The re- sults are very disappointing. Only about 50 per cent show a reasonable agreement, except in the case of Sample 3 where about 70 per cent are fair. Activity of water-insoluble organic nitrogen (Jones method).—Only about 50 per cent of the results show a reasonable agreement. Permanganate-insoluble organic nitrogen (Street method).—These results are also rather disappointing. As usual, however, there is less variation than in the Jones method for the liberated nitrogen. The majority of the results on Sample 1 are too high, indicating incomplete washing of the permanganate residue. About one-third of the results on Samples 2 214 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 and 4 are also too high. The results are quite satisfactory on Sample 3 and, omitting Spears, highly satisfactory on Sample 5, the complete fertilizer. Activity of water-insoluble organic nitrogen (Street method)—These results show a much greater uniformity and closer agreement than do the results obtained by the Jones method, as has been the case generally in comparative work on these two methods. ZINC-FERROUS SULPHATE-SODA METHOD. TABLE 2.—SAMPLE 6. In most cases there is little difference in the results whether a 0.5 or 5 gram sample be weighed, made to volume and a 0.5 gram aliquot taken. In the table the results of Ingham should be rejected, as they were obtained by the modified Gunning method. If the two results over 16 per cent also be rejected and the remainder of the results in the column headed “0.5 gram sample”’ alone be considered, these nine results fall into three groups: (1) five results, 15.23—15.39, average 15.30; (2) two results 15.50 and 15.60, average 15.55; and (3) two results, 15.70 and 15.84, average 15.77. The first average differs from the second by 0.25 and from the third by 0.47 and the second average from the third by 0.22 per cent. While individual workers, who reported more than one result (and most of them did)obtained very closely agreeing results, the referee does not think the results as a whole warrant the adoption of this method as official without further work. The opinion of the referee is further strengthened by the fact that, whereas more than half of the results fall in the first group (1), R. E. Ingham obtained by the modified Gunning method 15.68 and 15.64, by the official Kjeldahl method! 15.60, 15.72 and 15.68, by the official Gunning method? 15.82, 15.73 and 15.78, and C. F. Inman obtained 15.70 by both the Kjeldahl- Gunning with oxid of mercury and with copper sulphate, which would indicate that the results reported in groups (2) and (3) are the most correct. TOTAL NITROGEN. KJELDAHL-GUNNING-ARNOLD METHOD WITH OXID OF MERCURY AND WITH COPPER SULPHATE. The results are quite satisfactory with both of these methods, with good agreement, but a slight tendency to higher results with oxid of mercury. 1J. Assoc. Official Agr. Chemists, 1916, 1: No. 4 (ID), 5. 2Tbid., 7. 1919] BRACKETT AND HASKINS: NITROGEN 215, TABLE 3. Comparative results K jeldahl-Gunning-Arnold Method with orid of mercury and with copper sulphate. TOTAL GOOD RESULTS NUMBER | MAXIMUM METHOD OF DIFFER- AVERAGE RESULTS ENCE NUMBER MINIMUM MAXIMUM SAMPLE 1 DRIED BLOOD | per cent per cent per cent per cent Kjeldahl-Gunning-Arnold wathycopper ..\s./s.... b/sc6 13 0.53 7 12.65 12.80 12.75 Kjeldahl-Gunning-Arnold ‘ with mercury.......... 11 0.71 7 12.72 12.88 12.80 SAMPLE 2 TARTAR POMACE AND | DRIED BLOOD Kjeldahl-Gunning-Arnold with copper............ 13 0.40 9 5.38 5.50 5.44 Kjeldahl-Gunning-Arnold with mercury.......... 11 0.39 7 5.38 5.58 5.44 SAMPLE 3 NITROLENE K jeldahl-Gunning-Arnold 2 with copper............ 13 0.46 10 7.00 7.18 7.07 Kjeldabl-Gunning-Arnold with mercury.......... ll 0.40 8 7.02 7.26 7.11 SAMPLE 4 NITROGENOUS MANURE K jeldahl-Gunning-Arnold y Viculi ho) 0) (eee 13 0.47 8 8.12 8.32 8.23 K jeldahl-Gunning-Arnold with mercury.......... 11 0.42 8 8.26 8.44 8.34 SAMPLE 5 MIXED FERTILIZER Kjeldahl-Gunning-Arnold WAGE COPPET «.. 2c. aes Ja.01s 13 0.36 11 3.36 3.57 3.45 Kjeldahl-Gunning-Arnold with mercury.......... 10 0.30 Zi 3.36 | 3.57 3.46 CONCLUSIONS. (1) In the zine-ferrous sulphate-soda method for nitrates, the results of the different workers are too variable. The chief difficulty in the ma- nipulation of the method lies in the distillation with the use of glass wool in the neck of the flask. Further work should be done on this method without the use of the glass wool, with special reference to the inclination of the flask in the distillation, the use of a flask with a long neck, and the rate of distillation. It may be found, as Mr. B. F. Robert- son, of South Carolina, stated last year', that though this method has 1 J. Assoc. Official Agr. Chemists, 1915, 1: No. 3, 390. 216 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 the merit of rapidity, it requires too close attention to details where a large number of determinations have to be carried on at the same time. Certainly the method needs and appears to warrant further investigation. (2) Further work is necessary on the determination of the water- insoluble organic nitrogen. A better agreement this year between the results of the workers in different laboratories is probably due to the use of the Gunning-copper method, but perhaps more to the use by all workers of 1 gram instead of different amounts for the Jones and for the Street method. Sample contained nitrate. © Determination by plain Kjeldahl process. 1 Z. anal. Chem., 1883, 22: 375. 222 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 Nitrogen determinations with or without permanganate (by Scovell* salicylic acid modifica- tion unless otherwise noted).—Continued. NITROGEN PERCENTAGES TIME oO SAMPLE ANALYST areas: WITHOUT! WITH Loss NOTES REGARDING METHOD TION PERMAN- | PERMAN- WITH (HOURS) GANATE GANATE | PERMAN- GANATE SERIES II. 3258” | H. D. Edmiston. . Pa 3.65 3.45 0.20 | In Series IL when per- manganate was used, 3271 | H. D. Edmiston. . 23 2.52 2.18 0.34 its addition was begun instantly after the 3309 | H. D. Edmiston. . 24 2.1 1.81 0.30 flame was extinguish- ed, but was made avery 3324 | H. D. Edmiston. . 23 2.04 1.71 0.33 little at a time. The amount added was 0.5 3506" | Walter Thomas. . 23 2.03 1.83 0.20 gram, except in the case of the last two 3505 | Walter Thomas. . 23 1.39 0.64 trials of Sample 3506, when it was 1.5 and 3.5 3506 | Walter Thomas. . 2F eee 1.48 0.55 grams, respectively. Mercury and sodium hyposulphite used. b Sample contained nitrate. SERIES III. 3478 | Walter Thomas. . 22 3.30 | 3.00(a)| 0.30 | In Series III permanga- 2.64 (b)| 0.66 nate, amount not re- corded, was not added in one case; in a second case (a) gradually; in a third case (b) at once. A single sample was used, and the ad- dition begun instantly upon extinguishing the flame. SERIES IV. 3259 | Walter Thomas..| 3 1.91 | 1.61 | 0.30 | In Series IV, 1 gram of permanganate was 3309 | Walter Thomes..| 3 | 2.23 | 1.91 | 0.32 | 2dded mstantly uae extinguishing the = , flame, to the first two 3442 | Walter Thomas. . 3 4.73 4.60 0.13 samples, all at once; to the second pair, © 3271 | Walter Thomas. . 3 2.62 2.32 0.30 gradually. 1919] FREAR, THOMAS AND EDMISTON: KJELDAHL METHOD 223 Nitrogen determinations with or without permanganate (by Scovell* salicylic acid modifica- tion unless otherwise noted).—Continued. NITROGEN PERCENTAGES TIME SAMPLE ANALYST ee WITHOUT WITH LOSS NOTES REGARDING METHOD TION PERMAN- | PERMAN- WITH (HOURS) GANATE GANATE PERMAN- GANATE SERIES V°. 3506 | Walter Thomas. . 23 2.10 |1.78(a)) 0.30 | 0.7 gram added during , 30 seconds, beginning 3506 | Walter Thomas. . 24 2.07 | 1.82(a)! 0.26 instantly after flame was extinguished. 3506 | Walter Thomas. . aE 2.06 |1.90(b)| 0.18 | 0.6 gram added as in (a). 3506 | Walter Thomas. . 25 2.08 | 1.99 (c)| 0.09 | 0.3 gram added as in (a). 3506 | Walter Thomas. . 23 2.10 | 2.01 (d)| 0.07 | 0.2 gram added during 30 seconds, but beginning SVCD Epes sells 212 ols 2.08 30 seconds after flame was extinguished. 3506 | Walter Thomas. . 23 1.98 (e)| 0.10 | 0.7 gram added during 30 seconds, but beginning 1 minute after removal of flame. 3506 | Walter Thomas. . 24 2.05 (f)} 0.03 | 0.2 gram added as in (e). 3506 | Walter Thomas. . 24 2.12 (g) 0.7 gram added during 30 seconds, but beginning 3506 | Walter Thomas. . 23 2.08 (g) 2 minutes after re- moval of flame. 3506 | Walter Thomas. . 2 1.82 (h)} 0.26 | 0.7 gram added in 4 sec- onds at once after re- moyal of flame. 3506 | Walter Thomas. . » 2.06 (i)| 0.02 | 0.7 gram added in 4 sec- onds, beginning 1 min- ute after removal of flame. 3506 | Walter Thomas. . 2 1.78 (j)| 0.30 | 1.0 gram added as in (i). 3506 | Walter Thomas. . 2 1.68 (k)| 0.40 | 1.0 gram added as in (h). 3506 | Walter Thomas. . 2 2.10 (1) 0.7 gram added as in (g). 3506 | Walter Thomas. . 2 1.90(m)|} 0.18 | 0.7 gram added during 13 minutes beginning instantly after remoy- al of flame. ® Mercury and sodium hyposulphite used. : > Determinations of Series VY were confined to a single sample. 224 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 Nitrogen determinations with or without permanganate (by Scovell* salicylic acid modifica- tion unless otherwise noled).—Continued. NITROGEN PERCENTAGES | TIME OF SAMPLE ANALYST DIGES- WITHOUT WITH | Loss NOTES REGARDING METHOD TION PERMAN- | PERMAN- WITH (HOURS) GANATE GANATE | PERMAN- GANATE sERvES v°.—Continued. 3506 | Walter Thomas. - 2 .... |1.79(n)| 0.29 | 0.7 gram added in small quantities during di- gestion when liquid was brown; liquid digested 3 minutes after the ad- dition, until colorless. to 3506 | Walter Thomas. . 1.68 (o)| 0.40 | Like (m), but 3 grams ded. ad bo 3506 | Walter Thomas. . 2.06 (p)| 0.02 | 1.0 gram of lead peroxid substituted for per- manganate. ® Mercury and sodium hyposulphite used. b Determinations of Series V were confined to a single sample. These results show that for the considerable range of fertilizer mix- tures represented the addition of permanganate causes a distinct loss of nitrogen. The loss depends somewhat upon the amount of perman- ganate, but chiefly upon the time of addition. If the addition is delayed for two minutes after removal from the flame, no loss is observed. This implies that there is a critical temperature below which the reaction causing the loss does not take place. Four observations of temperature in the digestion flasks about the time of completion of digestion were made. For the plain Kjeldahl method 328°, 329°, 329° and 327°C. were observed. The modified Scovell method gave 344°, 345°, 342° and 345°C. Attempts were made to measure the rate of cooling but were unsuccessful. The fall is quite rapid—approximately 100°C. in two minutes after removal from flame. Kjeldahl, in the paper above mentioned, states that when perman- ganate is added the reaction is vigorous enough to cause evolution of light. No case of the appearance of luminosity was noted when perman- ganate was added at the instant of removal from the flame. This phe- nomenon occurred chiefly during the period from the fourth to the twelfth minute after removal from the flame. Kjeldahl’s mention of the occurrence of luminosity suggests that the time of its addition must have been longer after the removal from the flame than the adverb “imme- diately” would indicate. —— ee 1919) EWING: CHEMICAL REAGENTS 225 REPORT ON TESTING CHEMICAL REAGENTS. By C. O. Ewrne' (Bureau of Chemistry, Washington, D. C.), Referee. The subject of testing chemical reagents has been given some atten- tion by the Bureau of Chemistry and the results have fully justified the effort. Manufacturers know that only the better grades will be accepted and it is rare that lower grades are offered. The few exceptions, how- ever, are sufficient to justify the testing of all chemicals received. % It has been difficult to obtain absolute alcohol which will contain 99.8 per cent or more of absolute alcohol and 10 mils of which will not de- colorize 1 drop of 1:1,000 potassium permanganate in 10 minutes. This, in the opinion of the referee, is because the manufacturers do not take the heart of the distillate but try to stretch each run as far as possible. The Bureau of Chemistry now makes its own absolute alcohol which contains 99.9 to 99.95 per cent of absolute alcohol. Methyl alcohol has been rejected because of excessive residue on evaporation, the residue varying from 40 to 70 mg. per 100 cc. Difficulty has been experienced in obtaining amyl alcohol boiling be- tween 128° and 132°C. Amy] alcohol consists of varying proportions of iso-amyl-alcohol (b. p. 131°C.) and normal amyl alcohol (b. p. 128°C.). When poorly made it may contain lower boiling alcohols. If 90 per cent distils over between 128° and 132°C. it can be considered satis- factory. An excessive amount of residue was found in absolute ether preserved in cork-stoppered bottles. The stoppers have been protected and no further trouble experienced. One sample of benzol contained an appre- ciable amount of carbon disulphid. Few samples of animal charcoal complied with the specifications of the Bureau of Chemistry of “Not more than 4 per cent ash, and an extract with 3 per cent potassium hydroxid colored not deeper than light straw”. A new type of charcoal used in a number of the Bureau of Chemistry laboratories, known as “Ebonite’’, has given great satisfaction. Difficulty was experienced in obtaining satisfactory sodium silicate (40 per cent). Specifications were furnished the manufacturers as fol- lows: “The total sodium oxid to silicon dioxid should be not less than 40 per cent. The proportion of sodium oxid to silicon dioxid should be between the ratio of 1 :2.4 and 1:2.6.” A satisfactory product was then obtained. The following specifications were drawn up for asbestos: (1) The material should be prepared from pure long-fiber asbestos which has been shredded in such a manner as to leave the fibers loosely separated, varying in length 1 Present address, United Drug Company, Boston, Mass. 226 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 preferably from 1-2} cm. Fibers should be nearly pure white, with silky luster, capable of fine subdivision when shaken in water, and easily flexed without breaking. The material should possess good matting qualities. (2) It should be practically free from all impurities and contain not more than traces of iron. (A small fragment of asbestos heated to glowing should show on cooling no perceptible change in color and texture.) (3) The total loss after heating with 20% hydrochloric acid for 30 minutes and igniting at red heat 30 minutes should not exceed 3%. Five samples have been received under these specifications. Two complied with the requirements, one losing 3 per cent by the acid- ignition treatment and the other only 1.1 per cent. One sample of tin-foil was found to contain 97.7 per cent of lead. The Ninth Revision of the U. S. Pharmacopoeia permits the use of acetanilid as a preservative in hydrogen peroxid. A recent patent men- tions the use of cinchonidin. These substances should be guarded against in hydrogen peroxid for use as a reagent. Potassium hydroxid prepared by electrolysis is easily obtained and is of very good quality for most purposes but may contain up to 0.2 per cent of chlorid. The manufacturers have agreed to change their labels to indicate the presence of chlorid. A method for determination of alcohol was given which is essentially the same as subsequently published in the Ninth Revision of the U. S. Pharmacopoeia. RECOMMENDATIONS. It is reeommended— (1) (a) That chemical reagents labeled “U.S. P.” be tested according to the specifications of the U. S. Pharmacopoeia, Ninth Revision (1916). (b) (Tentative) That, unless otherwise specified, all other chemical reagents be tested according to Merck’s “Chemical Reagents, Their Purity and Tests”, second edition, 1914, by Henry Schenck. This pub- lication is a standard on the testing of chemical reagents. It represents the work of a great number of chemists, extending over many years. Rather than wait until such a comprehensive set of tests could be worked out by the members of this association, it would seem advis- able to adopt this latest revision of Merck. (2) That the method mentioned above be used for the determination of ethyl alcohol in pharmaceutical preparations. The methods of this association do not include a general method for alcohol in pharmaceutical preparations. The great diversity of alcoholic pharmaceuticals makes it extremely difficult to provide a method that will apply in every instance. The attempt has been made to prepare a general outline for procedure and then to give the most important and most frequently required modifications. The alcohol refractivity tables 1919} EWING: CHEMICAL REAGENTS 227 prepared by B. H. St. John have already been presented to the associa- tion at a previous session". (3) That the work on methods for the detection and determination of ethyl alcohol, methyl alcohol, amy! alcohol, and acetone be continued, together with work on immiscible organic solvents. The data at hand appear to warrant a continuation of the study of these chemicals. On motion of W. W. Skinner, the committee on recommendations of referees and revision of methods was instructed to prepare a form on which referees will be required to submit their several recommendations. The committee on editing methods of analysis submitted the following report on the definition of the terms “provisional”, ‘optional’, and “alternative’’: “A ‘provisional’ method is a method which has been reported by the appropriate committee upon recommendation of referees, published in the proceedings, and approved by the association, but which has not been tested sufficiently to warrant its adoption as an official method. “The terms ‘optional’ and ‘alternative’ have no place in the designa- tion of methods, and in the opinion of your committee should be eliminated.”’ Approved. The committee on amendment to the Constitution and By-Laws stated that more time would be necessary before a report could be presented. Mr. Ross moved that a vote of thanks be given the president, which was unanimously carried. The convention adjourned, at 4.20 P. M., to meet in Washington, D. C., at the call of the executive committee. 1 J. Assoc. Official Agr. Chemists, 1916, 2: No. 2 (II), 208-35. PROCEEDINGS OF THE THIRTY-THIRD ANNUAL CONVENTION OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS, 1916. OFFICERS, REFEREES, ASSOCIATE REFEREES AND COM- MITTEES OF THE ASSOCIATION OF OFFICIAL AGRI- CULTURAL CHEMISTS, FOR THE YEAR ENDING NOVEMBER, 1917. Honorary President. H. W. Wirey, Woodward Building, Washington, D. C. President. J. K. Haywoop, Bureau of Chemistry, Washington, D. C. Vice-President. P. F. Trowsringe, Agricultural Experiment Station, Agricultural College, N. Dak. Secrelary~Treasurer. C. L. Ausserc, Box 744, 11th Street Station, Washington, D. C. Additional Members of the Execulive Committee. B. B. Ross, Polytechnic Institute, Auburn, Ala. H. C. Lyruaor, State Department of Health, Boston, Mass. Referees. Phosphoric acid: W. J. Jones, jr., Agricultural Experiment Station, La Fayette, Ind. (deceased). Nitrogen: H. D. Haskins, Agricultural Experiment Station, Amherst, Mass. Potash: T. D. Jarrell, Bureau of Chemistry, Washington, D. C. Soils: C. B. Lipman, Agricultural Experiment Station, Berkeley, Cal. Inorganic plant constituents: J. ¥. Breazeale, Bureau of Plant Industry, Riverside, Cal. Insecticides and fungicides: O. B. Winter, Agricultural Experiment Station, E. Lansing, Mich. Water: J. W. Sale, Bureau of Chemistry, Washington, D. C. Foods and feeding stuffs: (Not appointed.) Dairy products: J. Hortvet, Old Capitol, St. Paul, Minn. Saccharine products: M. N. Straughn, Bureau of Chemistry, Washington, D. C. (deceased). 228 1919) REFEREES AND ASSOCIATE REFEREES 229 Drugs: W. O. Emery, Bureau of Chemistry, Washington, D. C. Testing chemical reagents: C. O. Ewing, United Drug Company, Boston, Mass. Micro-analylical methods: B. J. Howard, Bureau of Chemistry, Washington, D. C. Food preservatives: A. &. Seeker, U.S. Appraiser’s Stores, New York, N. Y. (deceased). Coloring matters in foods: W.E. Mathewson, Bureau of Chemistry, Washington, D. C. Melals in foods: David Klein, 1410 Kimball Building, Chicago, Hl. Fruil and fruit products: W. D. Bigelow, National Canners Association, Washington, D. C. M. N. Straughn, Bureau of Chemistry, Washington, D. C. (deceased). Canned vegetables: W. D. Bigelow, National Canners Association, Washington, D. C. Cereal foods: J. A. LeClerc, Miner-Hillard Milling Co., Wilkes-Barre, Pa. Wines: B. G. Hartmann, Transportation Building, Chicago, Ill. Soft drinks (bottlers’ products): W.W. Skinner, Bureau of Chemistry, Washington, D. C. Distilled liquors: J. 1. Palmore, Bureau of Chemistry, Washington, D. C. Beers: H.S. Paine, Bureau of Chemistry, Washington, D. C. Vinegars: W. A. Bender, Douglas Packing Co., Rochester, N. Y. Flavoring extracts: A. E. Paul, Transportation Building, Chicago, Ill. Meat and meat products: C. E. Marsh, State Department of Health, Boston, Mass. Edible fats and oils: R. H. Kerr, Bureau of Animal Industry, Washington, D. C. Spices and other condiments: H. E. Sindall, Austin Nichols & Co., Brooklyn, N. Y. Cacao products: E. Bloomberg, Pompeian Co., Baltimore, Md. Coffee: H. A. Lepper, Bureau of Chemistry, Washington, D. C. Tea: Miss E. A. Read, Bureau of Chemistry, Washington, D. C. Baking powder: H. ©. Patten, Bureau of Chemistry, Washington, D. C. Associale Referees. Phosphorie acid: Basic slag to cooperate with committee on vegetation tests on the availability of phosphoric acid in basic slag: 2. C. Shorey, 2706 Harrison Street, Wilmington, Del. Nilrogen: Special study of the Kjeldahl method: I. K. Phelps, Bureau of Chemistry, Wash- ington, D. C. Potash: J. T. Foy, Clemson College, S. C. Soils: Nitrogenous compounds: J. K. Plummer, Agricultural Experiment Station, Raleigh, N. C. Lime requirements: W. H. McIntire, Agricultural Experiment Station, Knoxville, Tenn. Inorganic plant constituents: (Not appointed.) Insecticides and fungicides: J. J. T. Graham, Bureau of Chemistry, Washington, D. C. Water: J. C. Diggs, State Board of Health, Indianapolis, Ind. Foods and feeding stuffs: Sugar: A. Hugh Bryan, Arbuckle Bros., Old Slip and Water Streets, New York, NS Yi Crude fiber: C. K. Francis, Transcontinental Oil Company, Tulsa, Okla. Stock feed adulteration: Miss B. H. Silberberg, Bureau of Chemistry, Washing- - ton, D.C. Organic and inorganic phosphorus: (Not appointed). Water: J. O. Clarke, U. S. Custom House, Savannah, Ga. 230 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 Dairy products: Separation of nitrogenous substances in milk and cheese: L. L. Van Slyke, Agri- cultural Experiment Station, Geneva, N. Y. Saccharine products: Maple products: J. F. Snell, Macdonald College, Quebec, Canada. Honey: S. F. Sherwood, Bureau of Plant Industry, Washington, D. C. Sugar house products: P. A. Yoder, Bureau of Plant Industry, Washington, D. C. Drugs: Medicinal plants: A. Viehoever, Bureau of Chemistry, Washington, D. C. Alkaloids: H. C. Fuller, Institute of Industrial Research, Washington, D. C. Synthetic products: C. D. Wright, Bureau of Chemistry, Washington, D. C. Balsams and gum resins: E. C. Merrill, United Drug Company, Boston, Mass. Enzyms: V. K. Chesnut, Bureau of Chemistry, Washington, D. C. Meat and meat products: Separation of nitrogenous compounds in meat products: P. F. Trowbridge, Agri- cultural Experiment Station, Agricultural College, N. Dak. Meat extracts: T. M. Price, Bureau of Animal Industry, Washington, D. C. PERMANENT COMMITTEES. Cooperation with Other Committees on Food Definitions. Wm. Frear (State College, Pa.). Chairman. Julius Hortvet, St. Paul, Minn. J. P. Street, Indianapolis, Ind. Recommendations of Referees and Revision of Methods. (Figures in parentheses refer to year in which appointment expires.) B. B. Ross (Auburn, Ala.), Chairman. SuscommitTTEeE A: A. J. Patten (1918), (Agricultural Experiment Station, E. Lansing, Mich.), Chairman, GC. C. McDonnell (1922), B. B. Ross (1920). (Phosphoric acid (basic slag to cooperate with committee on vegetation tests on the availability of phosphoric acid in basic slag), nitrogen (special study of the Kjeldahl method), potash, soils (nitrogenous compounds, lime requirements), inorganic plant con- stituents, insecticides and fungicides and water.) SuspcomMiTteEe B: R. E. Stallings, deceased, (1918), (Georgia Department of Agricul- ture, Atlanta, Ga.), Chairman, C. A. Browne (1922), H. C. Lythgoe (1920). (Foods and feeding stuffs (sugar, crude fiber, stock feed adulteration, organic and inorganic phosphorus, water), dairy products (separation of nitrogenous substances in milk and cheese), saccharine products (maple products, honey, sugar house products), drugs (medicinal plants, alkaloids, synthetic products, balsams and gum resins, enzyms), testing chemical reagents and micro-analytical methods.) SuscomMirtEE ©: L. M. Tolman (1918), (Wilson & Company, Chicago, Il.), Chairman, J. P. Street (1922), R. E. Doolittle (1920). (Food preservatives, coloring matters in foods, metals in foods, fruit and fruit products, canned vegetables, cereal foods, wines, soft drinks (bottlers’ products), distilled liquors, beers, vinegars, flavoring extracts, meat and meat products (separation of nitrogenous compounds in meat products, meat extracts), edible fats and oils, spices and other condiments, cacao products, coffee, tea, baking powder.) x 1919} MEMBERS AND VISITORS 231 SPECIAL COMMITTEES. Board of Editors. C. L. Alsberg (Bureau of Chemistry, Washington, D. C.), Chairman. E. F. Ladd (1917). R. E. Doolittle (1919). J. P. Street (1918). L. L. Van Slyke (1920). Editing Methods of Analysis (U.S. Bur. Chem. Bull. 107, Rev.). R. E. Doolittle (Transportation Building, Chicago, Ill.), Chairman. W. A. Withers. A. F. Seeker (deceased). J. P. Street. G. W. Hoover. B. L. Hartwell. Vegetation Tests on the Availability of Phosphoric Acid in Basic Slag. C. B. Williams (College of Agriculture and Mechanic Arts, West Raleigh, N. C.), Chairman. C. G. Hopkins. B. L. Hartwell. H. D. Haskins. J. A. Bizzell. Amendment to the Constitution and By-Laws. B. B. Ross (Polytechnic Institute, Auburn, Ala.), Chairman. C. L. Alsberg. H. D. Haskins. Committee on Methods of Sampling Fertilizers to Cooperate with a Similar Commitiee of the American Chemical Society. C. H. Jones (Agricultural Experiment Station, Burlington, Vt.), Chairman. W. J. Jones, jr. (deceased). B. F. Robertson. Committee on Revision of Methods of Soil Analysis. C. B. Lipman (Agricultural Experiment Station, Berkeley, Cal.), Chairman. W. H. McIntire. A. W. Blair. E. C. Shorey. R. Stewart. MEMBERS AND VISITORS PRESENT. Abbott, J. S., Bureau of Chemistry, Washington, D. C. Ageton, C. N., Central Experiment Station, Santiago de las Vegas, Habana, Cuba. Albright, A. R., Jackson Laboratory, E. I. Du Pont Co., Wilmington, Del. Allen, Mrs. Mary S., Bureau of Chemistry, Washington, D. C. Allen, R. M., Ward Baking Co., New York, N. Y. Almy, Lloyd H., 1833 Chestnut Street, Philadelphia, Pa. Alsberg, C. L., Bureau of Chemistry, Washington, D. C. Appleman, Chas. O., College Park, Md. Badger, C. H., Bureau of Chemistry, Washington, D. C. Bailey, H. S., E. I. Du Pont Co., Wilmington, Del. Bailey, L. H., Bureau of Chemistry, Washington, D. C. Baker, E. L., Patent Cereals Co., Geneva, N. Y. Balcom, R. W., Bureau of Chemistry, Washington, D. C. 232 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 2 Barnard, H. E., American Institute of Baking, Minneapolis, Minn. Bartlett, G. M., Joseph Campbell Co., Camden, N. J. Bartlett, J. M., Agricultural Experiment Station, Orono, Me. Baughman, W. F., Bureau of Chemistry, Washington, D. C. Bell, Chas. E., State Department of Agriculture, Raleigh, N. C. Bennett, Miss B. M., Bureau of Chemistry, Washington, D. C. Beyer, G. F., Bureau of Internal Revenue, Washington, D. C. Bidwell, G. L., Bureau of Chemistry, Washington, D. C. Bigelow, W. D., National Canners Association, 1739 H Street, N. W., Washington, D. C. Birckner, Victor, Bureau of Chemistry, Washington, D. C. Bitting, A. W., National Canners Association, 1739 H Street, N. W., Washington, D. C. Bizzell, J. A., State College of Agriculture, Ithaca, N. Y. Blanck, F. G., State Department of Health, 16 W. Saratoga Street, Baltimore, Md. Blumenthal, P. L., Agricultural Experiment Station, Lexington, Ky. Borden, N. H., Bureau of Chemistry, Washington, D. CG. Bowling, J. D., jr., State College, College Park, Md. Boyle, Martin, Bureau of Chemistry, Washington, D. C. Boyles, F. M., McCormick & Co., Baltimore, Md. Brackett, R. N., Clemson Agricultural College, Clemson College, 5. CG. Bradbury, CG. M., State Department of Agriculture, Richmond, Va. Brattain, P. H., Corby Co., Washington, D. C. Breckenridge, J. E., American Agricultural Chemical Co., New York, N. Y. Brewster, J. F., Bureau of Chemistry, Washington, D. C. Brinton, C. S., Food and Drug Inspection Station, U. S. Appraiser’s Stores, Philadel- phia, Pa. Broughton, L. B., College Park, Md. Brown, B. E., Bureau of Plant Industry, Washington, D. C. Brown, H. H., Bureau of Chemistry, Washington, D. C. Bryan, T. J., Calumet Baking Powder Co., Chicago, Ill. Burnet, W. C., Bureau of Chemistry, Washington, D. C. (deceased). Campbell, W. G., Bureau of Chemistry, Washington, D. C. Carpenter, F. B., Virginia-Carolina Chemical Co., Richmond, Va. Carroll, John S., German Kali Works, Atlanta, Ga. Cathcart, G. S., Agricultural Experiment Station, New Brunswick, N. J. Chapin, R. N., Bureau of Animal Industry, Washington, D. C. Chesnut, V. K., Bureau of Chemistry, Washington, D. C. Child, Ernest, H. Reeve Angel & Co., 120 Liberty Street, New York, N. Y. Chittick, J. R., Jaques Mfg. Co., Chicago, Il. Clark, A. J., Agricultural College, E. Lansing, Mich. Clarke, J. 0., Food and Drug Inspection Station, U. S$. Custom House, Savannah, Ga. Cochran, G. B., State Dairy and Food Bureau, W. Chester, Pa. Collins, W. D., Bureau of Chemistry, Washington, D. C. Conner, S. D., La Fayette, Ind. Cook, F. G., Bureau of Chemistry, Washington, D. C. Cross, L. J., Cornell University, Ithaca, N. Y. Custis, H. H., Bureau of Animal Industry, Washington, D. C. Darlington, Homer T., Metallurgical Engineer, Natrona, Pa. Daudt, H. W., Jackson Laboratory, E. I. Du Pont Co., Wilmington, Del. Davidson, J., Bureau of Chemistry, Washington, D. C. 1919] MEMBERS AND VISITORS 233 Davis, R. O. E., Bureau of Soils, Washington, D. C. Deemer, R. B., Bureau of Plant Industry, Washington, D. C. Dodge, C. O., Food and Drug Inspection Station, U. S. Appraiser’s Stores, New York, N. Y. Donk, P. J., National Canners Association, 1739 H Street, N. W., Washington, D. C. Doolittle, R. E., Food and Drug Inspection Station, Transportation Building, Chicago, il. Doran, J. M., Bureau of Internal Revenue, Washington, D. C. Doyle, Miss A. M., Room 9114 Du Pont Building, Wilmington, Del. Dubois, W. L., 215 Galene Street, Milwaukee, Wis. Dunbar, P. B., Bureau of Chemistry, Washington, D. C. Eimer, W. R., Eimer & Amend, New York, N. Y. Eldred, F. H., General Chemical Co., Laurelville, N. Y. Ellett, W. B., Agricultural Experiment Station, Blacksburg, Va. Emery, J. A., Bureau of Animal Industry, Washington, D. C. Emery, W. O., Bureau of Chemistry, Washington, D. C. Emmons, F. W., Washburn-Crosby Co., Minneapolis, Minn. Enslow, L. H., Miraflores Filtration Plant, Ancon, Panama, ©. Z. Eoff, J. R., jr., Bureau of Internal Revenue, Washington, D. C. Eskew, H. L., State Food and Drug Department, Nashville, Tenn. Ewing, C. O., United Drug Co., Boston, Mass. Fairchild, J. G., Bureau of Chemistry, Washington, D. C. Ferris, L. W., Bureau of Chemistry, Washington, D. C. Fitzgerald, F. F., National Canners Association, 1739 H Street, N. W., Washington, DG: Fowler, G. Scott, State Department of Agriculture, Atlanta, Ga. Fraps, G. S., Agricultural Experiment Station, College Station, Texas. Frary, G. G., State Food and Drug Commission, Vermilion, S. Dak. Frear, Wm., State Department of Agriculture, State College, Pa. French, D. M., Alexandria Fertilizer & Chemical Co., Alexandria, Va. Frey, R. W., John W. Heald & Co., Lynchburg, Va. Fry, Wm. H., Bureau of Soils, Washington, D. C. Fuller, A. V., Bureau of Animal Industry, Washington, D. C. Fuller, H. C., Institute of Industrial Research, Washington, D. C. Furber, F. B., Lederle Laboratories, New York, N. Y. Garby, C. D., Bureau of Chemistry, Washington, D. C. Gardiner, R. F., Bureau of Soils, Washington, D. C. Geiger, G. A., 43 Gaston Street, West Orange, N. J. Gibbs, H. D., E. I. Du Pont Co., Wilmington, Del. Gillespie, L. J., Bureau of Plant Industry, Washington, D. C. Goodrich, CG. E., Musher & Co., Baltimore, Md. Gore, H. C., Bureau of Chemistry, Washington, D. C. Gowen, P. L., Food and Drug Inspection Station, Park Avenue Building, Baltimore, Md. Grab, Eugene G., National Fruit Product Co., Woodward Building, Washington, D. C. Graham, J. J. T., Bureau of Chemistry, Washington, D. C. Grant, D. H., Bureau of Chemistry, Washington, D. C. Griffin, E. L., Bureau of Chemistry, Washington, D. C. 234 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. ITI, No. 2 Hand, W. F., Agricultural and Mechanical College, Agricultural College, Miss. Harris, H. L., Pacific Coast Borax Co., 100 William Street, New York, N. Y. Harrison, C. W., Food and Drug Inspection Station, Park Avenue Building, Baltimore, Md. Hartmann, B. G., Food and Drug Inspection Station, Transportation Building, Chicago, Ill. Hartwell, B. L., Agricultural Experiment Station, Kingston, R. I. Hausknecht, V. B., State Department of Agriculture, Harrisburg, Pa. Haywood, J. K., Insecticide and Fungicide Board, Washington, D. C. Haywood, W. G., State Department of Agriculture, Raleigh, N. C. Hazen, Wm., Bureau of Soils, Washington, D. C. Hellmuth, Everett A., Southern Railroad, Alexandria, Va. Herff, B. von, McCormick Building, Chicago, Ill. Hoagland, R., Bureau of Animal Industry, Washington, D. C. Holland, E. B., Agricultural Experiment Station, Amherst, Mass. Holman, H. P., Bureau of Chemistry, Washington, D. C. Holmes, A. D., Jackson Laboratory, E. [. Du Pont Co., Wilmington, Del. Hoover, G. W., Food and Drug Inspection Station, Transportation Building, Chicago, Tl. Hortvet, Julius, State Dairy and Food Department, Old Capitol, St. Paul, Minn. Houghton, H. W., Bureau of Chemistry, Washington, D. C. Howard, B. J., Bureau of Chemistry, Washington, D. C. Howes, C. C., Davison Chemical Co., Baltimore, Md. Hoyt, L. F., Larkin Co., Buffalo, N. Y. Hubbard, W. S., Food and Drug Inspection Station, U. 5. Appraiser’s Stores, New BY Ork; JIN’. ek. Hurst, L. A., Department of Agriculture, Washington, D. C. Huston, H. A., German Kali Works, 42 Broadway, New York, N. Y. Irwin, W. H., Swift & Co., Chicago, Il. Jacobs, B. R., Bureau of Chemistry, Washington, D. C. Jacobs, Mrs. B. R., 1502 Meridian Place, N. W., Washington, D. C. Jarrell, T. D., Bureau of Chemistry, Washington, D. C. Johns, C. O., Bureau of Chemistry, Washington, D. C. Jones, C. H., Agricultural Experiment Station, Burlington, Vt. Jones, D. B., Bureau of Chemistry, Washington, D. C. Jones, W. J., jr., Agricultural Experiment Station, La Fayette, Ind. (deceased). Jones, W. P., 3518 Newark Street, Cleveland Park, Washington, D. C. Keister, J. T., Bureau of Chemistry, Washington, D. C. Kellogg, J. W., State Department of Agriculture, Harrisburg, Pa. Kerr, R. H., Bureau of Animal Industry, Washington, D. C. Klein, David, State Food Department, 1410 Kimball Building, Chicago, Il. Klueter, Harry, Dairy and Food Commission, Madison, Wis. Knight, H. L., States Relations Service, Washington, D. C. Knight, O. D., Insecticide and Fungicide Board, Washington, D. C. Kunke, Wm. F., Bureau of Chemistry, Washington, D. C. LaForge, F. B., Bureau of Chemistry, Washington, D. C. Lang, H. L., States Relations Service, Washington, D. C. 1919] MEMBERS AND VISITORS 235 Lathrop, E. C., Jackson Laboratory, E. I. Du Pont Co., Wilmington, Del. Law, Thos. C., The Picard Law Co., Atlanta, Ga. LeClerc, J. A., Miner-Hillard Milling Co., Wilkes-Barre, Pa. LeCompie, T. R., Department of Agriculture, Washington, D. C. Lee, R. E., Fleischmann Co., Peekskill, N. Y. Lepper, H. A., Bureau of Chemistry, Washington, D. C. Liepsner, F. W., 825 Gravier Street, New Orleans, La. Linder, W. V., Bureau of Internal Revenue, Washington, D. C. Lipman, C. B., University of California, Berkeley, Cal. Lodge, F. S., Armour Fertilizer Works, Chicago, III. Lynch, Wm. D., Bureau of Chemistry, Washington, D. C. Lythgoe, H. C., State Department of Health, Boston, Mass. McClelland, Byron, Bureau of Chemistry, Washington, D. C. McCune, J. S., State Board of Health, Columbus, Ohio. McDonnel, M. E., Pa. R. R. Go., Altoona, Pa. McDonnell, C. C., Bureau of Chemistry, Washington, D. C. McDonnell, H. B., Agricultural Experiment Station, College Park, Md. McGeorge, W. T., Food and Drug Inspection Station, U. S. Appraiser’s Stores, San Francisco, Cal. MelIntire, W. H., Agricultural Experiment Station, Knoxville, Tenn. Mains, G. H., Massachuesetts Institute of Technology, Cambridge, Mass. Mallory, G. E., Bureau of Internal Revenue, Washington, D. C. Mason, G. I’., H. J. Heinz Co., Pittsburgh, Pa. Mastin, M. G., Porto Rico Fruit Exchange, San Juan, P. R. Mathewson, W. E., Bureau of Chemistry, Washington, D. C. Merrill, E. C., 77 Gainsborough Street, Boston, Mass. Middleton, Ellis S., Berwyn, Md. Miller, C. O., State Department of Health, 16 W. Saratoga Street, Baltimore, Md. Miller, H. M., National Canners Association, Los Angeles, Cal. Mitchell, A. S., Bureau of Chemistry, Washington, D. C. Morgan, W. J., Bureau of Chemistry, Washington, D. C. Mory, A. V. H., Sears, Roebuck & Co., Chicago, II. Murray, A. G., Bureau of Chemistry, Washington, D. C. Nealon, E. J., Bureau of Chemistry, Washington, D. C. Nelson, E. K., Bureau of Chemistry, Washington, D. C. Nollau, E. H., Experiment Station Record, Department of Agriculture, Washington, D.C. Nutt, G. S., Bureau of Internal Revenue, Washington, D. C. Obst, Mrs. Maud Mason, Food Research Laboratory, 1833 Chestnut Street, Philadel- phia, Pa. Paine, H. S., Bureau of Chemistry, Washington, D. C. Palmer, H. E., Bureau of Chemistry, Washington, D. C. Palmore, J. I., Bureau of Chemistry, Washington, D. C. Parkins, J. H., Royster Guano Co., Norfolk, Va. Parkinson, Miss N. A., Bureau of Chemistry, Washington, D. C. Patten, H. E., Bureau of Chemistry, Washington, D. C. Patterson, H. J., Agricultural Experiment Station, College Park, Md. 236 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 2 Phelps, I. K., Bureau of Chemistry, Washington, D. C. Phelps, Mrs. I. K., 1410 M Street, N. W., Washington, D. C. Pingree, M. H., American Agricultural Chemical Co., Baltimore, Md. Plummer, J. K., Agricultural Experiment Station, Raleigh, N. C. Pope, W. B., Bureau of Soils, Washington, D. C. Porch, M. B., H. J. Heinz Co., Pittsburgh, Pa. Power, F. B., Bureau of Chemistry, Washington, D. C. Rabak, Frank, Bureau of Plant Industry, Washington, D. C. Rabinowitz, Benjamin, Alart & McGuire Co., Brooklyn, N. Y. Randall, W. W., State Department of Health, 16 W. Saratoga Street, Baltimore, Md. Rather, J. B., Standard Oil Company, Chemical Laboratory, Brooklyn, N. Y. Reed, J. B., Bureau of Chemistry, Washington, D. C. Remsburg, C. G., State College of Agriculture, College Park, Md. Roark, R. C., General Chemical Co., Baltimore Md. Robb, J. B., State Department of Agriculture, Richmond, Va. Robinson, ©. H., Experimental Farm, Ottawa, Canada. Rodes, Wm., Agricultural Experiment Station, Lexington, Ky. Rogers, J. S., Kistler-Lesh Co., Morganton, N. C. Rogers, L. A., Bureau of Animal Industry, Washington, D.C. Rose, RB. E., State Department of Agriculture, Tallahassee. Fla. Ross, B. B., Polytechnic Institute, Auburn, Ala. Ross. 8. H., Cudahy Packing Co., East Chicago, Ind. Roth, G. B., Hygienic Laboratory, Washington, D. C. Rudnick, P., Armour & Co., Chicago, Tl. Runkel, Homer, Bureau of Chemistry, Washington, D. C. Runyan, E. G., Hutchins Building, Washington, D. C. Sale, J. W., Bureau of Chemistry, Washington, D. C. Sample, J. W., Department of Food and Drugs, Nashville, Tenn. Sarnuels, J. A., 133 17th Street, N. W., Washington, D. C. Savage, Mrs. G. O., 4212 Keokuk Street, N. W., Washington, D. C. Schreiner, Oswald, Bureau of Plant Industry, Washington, D. C. Schroeder, J. P., Bureau of Soils, Washington, D. C. Schwartze, E. W., Bureau of Chemistry, Washington, D. C. Seeker, A. F., Food and Drug Inspection Station, U.S. Appraiser’s Stores, New York, N. Y. (deceased). Seidell, Atherton, Hygienic Laboratory, Washington, D. C. Sherwood, 8S. F., Bureau of Plant Industry, Washington, D. C. Shorey, E. C., 2706 W. Harrison Street, Wilmington, Del. Shrader, J. H., Bureau of Plant Industry, Washington, D. C. Sievers, A. F., Bureau of Plant Industry, Washington, D. C. Silberberg, Miss B. H., Bureau of Chemistry, Washington, D. C. Simpson, G. E., Yale University, New Haven, Conn. Sive, Benj. E., Bureau of Standards, Washington, D. C. Skinner, J. J., Bureau of Plant Industry, Washington, D. C. Skinner, W. W.. Bureau of Chemistry, Washington, D. C. Smalley, Ff. N., Southern Cotton Oil Co., Savannah, Ga. Smith, ©. M., Bureau of Chemistry, Washington, D. C. Smith, H. R., Food and Drug Inspection Station, Park Avenue Building, Baltimore, Md. Smith, Joseph G., Bureau of Soils, Washington, D. C. 1919] MEMBERS AND VISITORS 237 Smither, F. W., Bureau of Standards, Washington, D. C. Snyder, C. F., Bureau of Standards, Washington, D. C. Snyder, Harry, Russell Miller Milling Co., Minneapolis, Minn. Spencer, G. C., Bureau of Soils, Washington, D. C. Stallings, R. E., State Department of Agriculture, Atlanta, Ga. (deceased). Steinkoenig, L. A., Bureau of Soils, Washington, D. C. Stephenson, Chas. H., Bureau of Chemistry, Washington, D. C. Stewart, Robert, University of [linois, Champaign, II. Street, J. P., 405 Indiana Trust Building, Indianapolis, Ind. Sutton, C. G., B. B. Culture Laboratory, 176 Palisade Avenue, Yonkers, N. Y. Taber, W. C., Food and Drug Inspection Station, Park Avenue Building, Baltimore, Md. Taistra, Theo., Eimer & Amend, New York, N. Y. Taylor, A. E., Bureau of Chemistry, Washington, D. C. Taylor, G. B., Bureau of Animal Industry, Washington, D. C. Taylor, J. N., Bureau of Animal Industry, Washington, D. C. Thatcher, R. W.. Agricultural Experiment Station, St. Paul, Minn. Thom, Chas., Bureau of Chemistry, Washington, D. C. Thomas, CG. C., Williams & Wilkins Co., Baltimore, Md. Thompson, Miss A. R., Bureau of Chemistry, Washington, D. C. Todd, A. R., Dairy and Food Department, Lansing, Mich. Toll, J. D., The American Fertilizer, Philadelphia, Pa. Treuthardt, E. L. P., Purchase and Storage, Subsistence Division, Inspection Branch, Munitions Building, Washington, D. C. Trowbridge, P. F., Agricultural Experiment Station, Agricultural College, N. Dak. Valaer, Peter, jr., Bureau of Internal Revenue, Washington, D. CG. Van Slyke, L. L., Agricultural Experiment Station, Geneva, N. Y. Veitch, F. P., Bureau of Chemistry, Washington, D. C. Vollertsen, J. J.. Morris & Co., Chicago, Ill. Vrooman, Carl S., Wilmington, III. Waldraff, P. H., General Chemical Co., Hudson Heights, N. J. Walker, L. S., Agricultural Experiment Station, Amherst, Mass. Walker, P. H., Bureau of Standards, Washington, D. C. Walters, E. H., Bureau of Plant Industry, Washington, D. C. Weatherhead, D. L., Food and Drug Department, Nashville, Tenn. Weber, F. C., Bureau of Chemistry, Washington, D. C. Weems, J. B., State Department of Agriculture, Richmond, Va. Wessling, Miss H. L., States Relations Service, South, Washington, D. C. White, H. J., State College, College Park, Md. Wilbert, M. I., Hygienic Laboratory, Washington, D. C. Wilcox, B. B., Food and Drug Inspection Station, Custom House, Savannah, Ga. Wiley, H. W., Woodward Building, Washington, D. C. Wiley, H. W., jr., Woodward Building, Washington, D. C. iley, Samuel W., Wiley & Co., Inc., 7S. Gay Street, Baltimore, Md. Wilson, J. B., Bureau of Chemistry, Washington, D. C. Wilson, S. M., Baugh Chemical Co., Baltimore, Md. Wise, L. E., Jackson Laboratory, E. I. Du Pont Co., Wilmington, Del. Withers, W. A., College of Agriculture and Mechanic Arts, Raleigh, N. C. Woodward, H. E., E. 1. Du Pont Co., Wilmington, Del. 238 ASSOCIATION OF OFFICIAL AGRIGULTURAL CHEMISTS [Vol. ITI, No. 2 PRESIDENT’S ADDRESS. THE PAST, PRESENT AND FUTURE OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS. By R. N. Brackerr (Clemson Agricultural College, Clemson College, S$. C.), President. MEMBERS OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS: In casting about for a suitable subject on which to address you, I followed, in all probability, the custom of nearly all past presidents, of reviewing the proceedings of the association, in order to find out the established precedents. I was at first rather surprised to find that my files of the proceedings began with the sixth annual meeting, 1889. Upon reaching the proceedings of the sixteenth annual meeting, I found that our much beloved charter member of this association, its third president, and its secretary for twenty-five years or more, to whom this association owes more than perhaps to any other one man, and who is still identified with us as honorary president, Dr. Harvey W. Wiley, had preserved for us the origin, establishment, and an account of the first fifteen meetings in an “Historical Sketch of the Association of Official Agricultural Chemists’*. Dr. Wiley stated that he had prepared this sketch because the earlier reports were not available. The association certainly owes Dr. Wiley a debt of gratitude for having written this sketch, which I commend to all the younger members of this association, together with other interesting historical matter which they will find in the proceed- | ings of the sixteenth annual meeting. The net result of this review, as far as the addresses of the presidents _ are concerned, is that with one or two exceptions all of the presidents — have made addresses, and with only two or three exceptions the address has been on the work of the association. Had not the only president who really repudiated his address been from South Carolina, I think I would have been strongly tempted to follow his precedent, or rather his lead, since his action never became a precedent, as I am heartily in agree- ment with President H. J. Patterson, when in his address before this association at its twenty-ninth meeting, 1912, he said’: ‘Each year it becomes more and more difficult, or perhaps I would better say that it seems less and less necessary for the president to perform this duty, owing to the fact that the work of the association has already been so well — mapped out and covered and the merits and defects so thoroughly appreciated.’ I, too, have been impressed with the completeness and 1 Presented Tuesday, November 21, as special order of business for 11.30 o'clock. 2U.S. Div. Chem. Bull. 57, 16-49. 3U.S. Bur. Chem. Bull. 162, 109. es ae 1919] BRACKETT: PRESIDENTS ADDRESS 239 thoroughness with which the various presidents have covered the whole field of the association’s work, but also with how well they have culti- vated, or I might say, harvested the adjacent territory. For example, President Hopkins at the twenty-third meeting gave his notable address on “The Duty of Chemistry to Agriculture’!; at the twenty-fifth meet- ing, President Harry Snyder spoke on ‘“The Training of the Agricultural Chemist’’?; at the twenty-seventh meeting, President W. A. Withers addressed us on “The Teaching of Chemistry in American Agricultural Colleges’’*; President G. S. Fraps, at the thirtieth meeting discussed the advances made in agricultural chemistry*. So there you are! What is one to do? It appears that everything has got to such a fine state of division that an appropriate subject would now be “‘Colloid Chemistry” in its various applications. I only wish I were competent to handle that subject, but I can only recommend it to those who come after me. After considering the matter from various angles, I have decided to risk the criticism or even the opprobrium of the o!der members of the association, in order to try at least to stir in the newer and younger members, who now make up the majority of its membership, a greater pride and interest in its past achievements, a more vital and sympa- thetic interest in its present activities, and a patriotic desire, not only to witness, but to have a share in its future development—The Past, Present, and Future of the Association of Official Agricultural Chemists. I can see from your looks that this seems to you to be a very presumptu- ous theme. My answer is that it is matter of history that “All gall is divided into three parts”. It will, of course, be necessary for me, in order not to tax your patience too much, to touch some of the three phases of my subject somewhat lightly. It has been my privilege to have been an active member of this asso- ciation for only the past six years, although I attended a few sessions of the seventeenth meeting, 1900. This fact will, therefore, account for the statement I am about to make, that the review I have recently made of the proceedings of the association has been to me both enlightening and profitable, nay more, inspiring. The personnel of the membership, the faithfulness in attendance and the unselfish labors of the members, the large amount and excellence of the work which has been done, these and many other things, some of which I wish to revert to again, have excited in me a greater respect and admiration for and appreciation of the past achievements of this association, a more genuine and sym- pathetic interest in its present welfare, and a keener and more earnest Tl. Agr. Exp. Sta. Bull. 105 (1906). 2U.S. Bur. Chem. Bull. 122, 110-4. 3 Thid., 137, 91-7. ‘J. Assoc. Official Agr. Chemists, 1915, 1: No. 1 (1), 158-63. 240 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 desire to contribute in every possible way to its future development and enlarged usefulness. May it not be laid down as a general principle, that a familiarity with the history and development of any subject is in very large measure the real basis of one’s vital and sympathetic interest in it, and of the actual pleasure derived from working in its fields, as well as the inspira- tion for such efforts as one is led to put forth in trying to enlarge its boundaries? So strongly do we believe in this principle that all students at Clemson College who pursue the study of chemistry for three years, are required to study the history of chemistry for at least two terms of the third year. The wide appreciation and recognition of this principle is evidenced in the plan, now quite common, of introducing historical references and cuts of the more prominent scientists in the modern text- books of chemistry and physics especially. This, then, is my apology, if any be necessary, for the selection of my subject. To the newer and younger members of the association much that I shall say will be new. To the older members, should they be disposed to think they are being imposed upon, I have a story to tell. Those of you who are familiar with Dr. Wiley’s sketch will recall the following interesting facts with regard to the origin and establishment of the association: “The condition of agricultural chemical work in the United States in 1880 was a peculiar one. The few chemists who were engaged in agri- cultural research were acting in complete independence of each other in regard to methods of investigation and of research. Some of them were using the methods employed by German chemists, while others followed the instruction given by the French or English agricultural chemists. There was no unity of interest in the profession nor any common system of work. The condition of analytical work may be truly described as chaotic. The result of such condition is easily imagined. There was no standard of comparison or reference. Buyers and sellers were con- tinually wrangling over analyses, which, made by different men follow- ing different methods, did not agree. “The sellers’ chemists uniformly obtained higher results than the buyers’, and thus the door to litigation was constantly open. “Strange as it may seem, the first steps toward correcting this pitiable condition did not come from the Department of Agriculture at Wash- — ington, but from the department of agriculture of one of the States. It was through the Hon. J. T. Henderson, commissioner of agriculture of Georgia, and at the instigation of Mr. H. J. Redding, now (1899) director of the Georgia station, that the first step toward uniformity of action among agricultural chemists of the United States was taken.” 1919) BRACKETT: PRESIDENTS ADDRESS 241 In response to Mr. Henderson’s two calls, issued May 20 and July 1, 1880, the first convention of agricultural commissioners and chemists met in the library hall of the Department of Agriculture at Wash- ington, July 28, 1880, with twenty men present. Naturally this body of men consisted chiefly of the commissioners of agriculture, and of representatives of State boards of agriculture, State chemists, and pro- fessors of chemistry in State universities and agricultural colleges, from those States using large amounts of commercial fertilizers. The purpose of this meeting, as stated by Mr. Henderson in his first call for the convention, was to secure “‘such uniformity of method in determining by chemical analysis the percentage of valuable ingredients in commer- cial fertilizers as will give more uniform and hence more satisfactory results. This is especially desirable in determining ‘reverted’ phosphoric acid.” Among the men who attended this epoch-making meeting are the names of not a few whose names are household words in agricultural chemistry. I commend the reading of the list, which is given in Dr. Wiley’s sketch. The program outlined for discussion at this meeting included the estimation of soluble “reverted” and insoluble phosphoric acid, nitrogen and potash; and, if time allowed, the method of arriving at commercial valuations and the agricultural and commercial valuation of “reverted”’ phosphoric acid. The entire question of phosphoric acid was referred to a committee of five—Drs. C. U. Shepard, jr., C. A. Goessmann, G. A. Liebig, E. H. Jenkins, and T. R. Wolf. In like manner the question of the estimation of nitrogen and potash was referred to a committee of five—Drs. W. M. Habirshaw, P. B. Wilson, Arthur T. Neale, C. Elton Buck, and N. A. Pratt. These two committees were then given an opportunity to prepare their reports, which were adopted and referred to a committee of two, who should copy them, giving the necessary details of the methods proposed, and send a copy to each member of the convention. Since so many agricultural chemists were present, it was suggested that it would be well to effect a permanent organization to meet from time to time and discuss topics of interest to the profession, and accord- ingly the following resolution offered by Dr. C. A. Goessmann was adopted: “Resolved, that this convention form a section in the sub- division of chemistry in the American Association for the Advancement of Science, and that their next meeting be held in Boston during the regular meeting of the aforesaid association.” The second convention of agricultural chemists was held at Boston, August 27, 1880. At this meeting, as the result of a statement of Dr. F. W. Clarke, that there had been inaugurated a movement to have a section of chemistry in the American Association for the Advancement 242 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 of Science, the following resolution was adopted: “Resolved, that a committee of three be appointed by the chair (Dr. Goessmann) to take such steps as may be necessary for securing the formation of a permanent chemical section in the American Association for the Advancement of Science, and the establishment of a subsection of agricultural chemistry in such permanent section, should it be formed.” The reports of the committees on the estimation of phosphoric acid, nitrogen and potash were made and the methods recommended adopted for one year. The following important resolution was also adopted: “Resolved, that a committee of five be appointed by the chair to secure the cooperation and experimental research of agricultural chemists, to collect and examine the various published methods of fertilizer analysis, and to make a report at the next meeting of this convention; and that we individually pledge ourselves to conduct, for this committee, any experiments or tests which they may desire.”’ Messrs. S. W. Johnson, C. U. Shepard, jr., Peter Collier, W. O. Atwater, and G. C. Caldwell were appointed on this com- mittee. About twenty-five were present at this meeting, which ad- journed to meet at the same time and place with the American Asso- ciation in 1881, the chairman and secretary to hold over. The third convention of agricultural chemists was held at Cincinnati, August 18, 1881. The outstanding results of this meeting were that no complaints had been heard with regard to the analysis of fertilizers, except as to insoluble or “reverted’’ phosphoric acid; that the com- mittee of five last mentioned had no report; that some progress was being made toward the formation of a permanent section of chemistry in the American Association for the Advancement of Science; that a long and warm discussion was indulged in on the methods of estimating insoluble or ‘‘reverted’’ phosphoric acid, the commercial chemists favor- ing the ammonium oxalate and the agricultural chemists the ammonium citrate method. The discussion ended in the appointment of a com- mittee of seven to consider the best method, and its details, for determin- ing insoluble or “reverted” phosphoric acid by the use of ammonium oxalate—one member to be chosen by the chair, three by the president of the Chemical and Fertilizer Exchange, and three to be the State chemists in charge of fertilizer control present at the meeting, viz., Drs. C. W. Dabney, jr., N. W. Lord, and H. C. White. A committee — of five was also appointed to continue the investigation of the whole subject of the determination of insoluble or “reverted”’ phosphoric acid. The chair appointed on this committee Messrs. A. R. Ledoux, C. A. Goessmann, E. H. Jenkins, G. A. Liebig, and C. M. Stillwell. I have given this matter in some detail because this appears to have been the subject upon which this convention split, and we must admit, I think, that the method of determining insoluble phosphoric acid is still a ‘rock 1919) BRACKETT: PRESIDENTS ADDRESS 243 of offence”. Embodied in the published minutes of this meeting are the report of the committee of seven, and also a letter from Dr. S. W. Johnson on the citrate method. A large number of chemists were present at this meeting, and a partial list is given by Dr. Wiley in his sketch. The meeting adjourned to reconvene at the call of the chair, provided a meeting were necessary before their organization as a sub- section of the American Association for the Advancement of Science. In the words of Dr. Wiley, “After the adjournment of the Cincinnati meeting the interest in the collaboration of the agricultural chemists seemed to die out. There was a certain feeling of antipathy—perhaps it is not well to make it so strong as this, but a strong feeling of incon- eruity existed—between the trade chemists on the one hand and the official chemists on the other. It was an unvoiced sentiment pervading the organization to the effect that an association composed of trade chemists and official chemists contained elements of instability which would prevent it from ever becoming highly useful.’’ Nevertheless, after the lapse of three years, Mr. Henderson again called a meeting of the agricultural chemists which was held in the Senate Chamber of the Capitol at Atlanta, Ga., May 15, 1884. There were present at this meeting, which was the precursor of the organization meeting here today: Judge J. T. Henderson, Prof. H. W. Wiley, Prof. N. W. Lord, Dr. Charles W. Dabney, jr., Prof. William A. Noyes, Dr. Charles U. Shepard, jr., Dr. J. W. Gascoyne, Dr. Allard Memminger, Mr. Philip E. Chazal, Dr. E. H. Jenkins, Prof. S. W. Johnson, Prof. H. C. White, Dr. N. P. Pratt, Mr. G. W. Benson, Mr. N. A. Pratt, Prof. John A. Myers, Prof. J. A. Burns, Mr. L. H. Jones, jr., Mr. C. L. Allen, Mr. G. W. Davison, Prof. W. C. Stubbs, Dr. Arthur T. Neate, Mr. P. T. Pendleton, Mr. C. B. F. Lowe, Mr. Leroy Broun, jr., Mr. C. M. Strahan, Mr. A. F. Crowell, Mr. C. V. Petraeus, Mr. Theo. Schumann, Mr. P. J. Schumann. After the opening address by Judge Henderson, Dr. C. U. Shepard, jr., made an introductory address in which he said, “Again we are assembled to consider the inconsistencies of agricultural chemical analysis. That they pinch all of us there isno doubt.” Dr. Shepard insisted that the chemists having official connection with agricultural colleges and experiment sta- tions and boards of agriculture should take the initiative in establishing uniform methods of procedure. As a commercial chemist he expressed his entire willingness to collaborate with the official chemists in every possible way, and also to receive and examine samples which might be submitted for comparative analysis. At this meeting three committees were appointed to consider methods of analysis, as follows: “(1) On methods of determining phosphoric acid, consisting of Drs. S. W. Johnson, H. C. White, and W. C. Stubbs. 244 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 (2) On methods of determining nitrogen, consisting of Messrs. P. E. Chazal, A. T. Neale, and J. A. Myers. (3) On methods of determining potash, consisting of E. H. Jenkins, W. J. Gascoyne, and H. W. Wiley. “Papers were read by F. B. Dancy, on comparison of some methods of determining ‘reverted’ phosphoric acid; by Professor Johnson, on com- parison of the action of ammonium citrate solution on phosphates at different temperatures; by Dr. Shepard, on certain apparatus for the determination of reduced phosphoric acid; by Dr. Wiley, on the action of ammonium citrate on ground bone, and the Gooch crucible in phos- phate estimations; by Dr. Shepard, on the insufficiency of the ‘Wash- ington’ method when applied to acid phosphates which have lain long in pile; by Dr. Memminger, on determining ‘reverted’ phosphates by the use of ammonium oxalate; by Dr. Dabney, on the Ruffle and copper oxid methods of determining nitrogen; by C. B. F. Lowe, on the de- termination of nitrogen from nitrates by the soda-lime process; by Dr. Memminger, on methods of determining potash. “The reports of the three committees on methods of analysis were received, discussed, amended, and adopted. The bulletin of the pro- ceedings of the Atlanta convention practically forms the first in the series of bulletins containing the methods of analysis of the Association of Official Agricultural Chemists. [t was compiled on the plan which was followed for many years thereafter, of having the proceedings and methods contained in the same bulletin, which plan was only abandoned when the volume of matter to be considered was so great as to necessitate two separate publications. “The convention, when it adjourned, agreed to meet in connection with the American Association for the Advancement of Science, in Philadelphia, the following September. “The Philadelphia meeting was held September 8, 1884. Dr. E. H. Jenkins was appointed chairman and Dr. C. W. Dabney acted as sec- retary. A committee appointed at the Atlanta meeting to consider the advisability of organizing the association as a subsection of the American Association for the Advancement of Science, recommended the formation of two associations. “First. The Association of Agricultural Chemists to be entirely distinct from the American Association to which should be left the discussion of the methods of analysis, ete. “Second. The subsection of the American Association for the Ad- vancement of Science, to be open to all agricultural chemists for the purposes of investigation and discussion. “The unanimous opinion expressed in the discussion of this subject was that an organization entirely separate from the American Associa- 1919) BRACKETT: PRESIDENT’S ADDRESS 245 tion would best advance the objects of the convention. A committee, consisting of Messrs. H. C. White, E. H. Jenkins, P. E. Chazal, J. A. Myers, and H. W. Wiley, was appointed to consider the form of organiza- tion and instructed to report the following day.” “On September 9 a formal organization took place, the present name of the association was adopted, officers for the following year were elected, and the convention resolved into the first annual meeting of the Associa- tion of Official Agricultural Chemists. Committees on phosphoric acid, potash, and nitrogen were appointed, and methods for the determination of phosphoric acid and potash in commercial fertilizers adopted as the official methods of the association. “The bulletin containing the proceedings of this meeting consists of eight pages of printed matter, in which are given the list of officers elected, the constitution, and the methods of analysis adopted officially by the association.” The second annual meeting of the Association of Official Agricultural Chemists was held in the library of the Department of Agriculture, be- ginning September 1, 1885. The officers of this convention were: Presi- dent, S. W. Johnson; vice-president, H. C. White; secretary-treasurer, C. W. Dabney, jr.; executive committee, E. H. Jenkins and H. W. Wiley. In the absence of President Johnson, Vice-President White took the chair, and, Dr. Dabney also being absent, Mr. Chazal acted as secretary. Hon. Norman J. Colman, Commissioner of Agriculture, ad- dressed the members, dwelling on the objects of the association, declar- ing his hearty sympathy with them, and expressing the hope that the association would not limit its attention to uniform methods of fertilizer analysis, but would extend its work to general chemical analysis, and commended to its special attention the standard of purity of foods and methods of detecting adulteration. Reports from the various commit- tees appointed at the previous meeting were received, and several papers on subjects pertaining to the work of the association read. Officers elected for the ensuing year were: President, H. W. Wiley; vice- president, C. W. Dabney, jr.; secretary-treasurer, Clifford Richardson; executive committee, W. J. Gascoyne and H. A. Huston. By invitation of Commissioner Norman J. Colman the third annual meeting of the association was held at the Department of Agriculture, August 26 and 27, 1886, and through his courtesy the proceedings were published as Bulletin No. 12 of the Division of Chemistry. At the opening of the convention Dr. Wiley, the president, addressed the asso- ciation, reviewing the work accomplished, presenting in detail the methods of fertilizer analysis in use in foreign countries and comparing them with the methods adopted by the association. Dr. Wiley closed his address by recommending that the investigations of the association 246 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. ITI, No. 2 be extended over a wider range of subjects, and expressing the opinion that every problem connected with chemical agricultural analysis was a proper subject for discussion at meetings. He suggested the advisa- bility of appointing a committee to consider the propriety of revising the clause in the constitution which limited the membership and the scope of investigations. Then followed reports on phosphoric acid, potash, and nitrogen, together with various papers on each of the three subjects. For the first time methods were adopted as official for nitrogen determinations, it being decided that the association would recognize as official either the Ruffle, Kjeldahl, absolute or soda-lime methods, when carried out according to the working details to be supplied by the committee. These methods were therefore printed as official methods of the association. Officers elected for the following year were: Presi- dent, E. H. Jenkins; vice-president, P. E. Chazal; secretary-treasurer, Clifford Richardson; executive committee, H. W. Wiley and M. A. Scovell. At this meeting two additional committees were appointed, namely, on food stuffs and dairy products, and proper methods adopted for the work of each. The fourth annual convention of the association was held in Washing- ton, August 16, 17, and 18, 1887, in the library of the Department of Agriculture, with thirty-five members present and President Jenkins in the chair. In his address Dr. Jenkins urged particularly an amendment of the constitution so as to admit as members of the association chemists from agricultural colleges who exercised no official control over the analy- sis of fertilizers, soils, cattle foods, dairy products, and other materials connected with agricultural industry, and also the chemists from the Treasury Department. This amendment was adopted toward the close of the meeting. Then followed in their regular order the report of the committee on the analysis of cattle food, the report on dairy products, phosphoric acid, potash, and nitrogen. In addition to the five existing committees on analytical methods, two more were added at this meeting, namely, one on the analysis of fermented liquors and the other on methods of analysis of sugars. The fifth annual convention of the association was held in the library of the Department of Agriculture at Washington, August 9, 1888, with President Chazal in the chair. The other officers of this meeting were: vice-president, W. J. Gascoyne; secretary, Clifford Richardson; addi- tional members of the executive committee, John A. Myers and E. H. Jenkins. There occur in the minutes of this meeting three entries worthy of special note: First, that the president omitted the usual address in order to proceed with the business of the convention; second, that Prof. F. W. Clarke, of the Geological Survey, addressed the mem- bers of the association on the desirability of forming a national chemical 1919] BRACKETT: PRESIDENTS ADDRESS 247 society, and in harmony with this idea a committee, consisting of Messrs. H. W. Wiley, W. C. Stubbs, and E. A. de Schweinitz, was appointed to represent the association in any action which might be taken in this direction; third, that instead of a committee of three, as was the previous custom, it was decided by the association to put one person, called a Reporter, in charge of each of the subjects considered by the association. The sixth annual convention of the association was held in the Seed Division of the Department of Agriculture, beginning September 10, 1889, with President John A. Myers in the chair. The other officers of the meeting were: Vice-president, M. A. Scovell; secretary, Clifford Richardson; executive committee, H. W. Wiley and William Frear. The session was opened with an address by Hon. Edwin Willets, Assistant Secretary of Agriculture. President Myers followed with an address of some length, on the objects of the association, laying special stress on the necessity of such accuracy and reliability of methods as to compel respect for and trust in the conclusions of the association, citing statistics showing the influence of the association on commercial activity, and comparing its work with that of foreign nations having the same object in view. The feasibility of organizing a national chemical society was discussed at length and on motion the association approved the proposi- tion. Reports were received from the various committees and disposed of as seemed best. Mr. Richardson resigned the position of secretary and Dr. H. W. Wiley was chosen to fill the vacancy. Thirty-seven members and visitors were present. The seventh annual convention of the association was held at Wash- ington in the lecture hall of the National Museum, beginning August 28, 1890. with President M. A. Scovell in the chair. The other officers of the meeting were: Vice-president, G. C. Caldwell; secretary, H. W. Wiley; additional members of the executive committee, J. A. Myers and E. H. Jenkins. In his presidential address, Dr. Scovell devoted his time main!y to recommendations in regard to special lines of work of the association. The work on cattle foods had now become so ex- tended that, at this convention, it was deemed necessary to divide the work between two reporters, one to consider foods high in carbohydrates, and the other to devote himself to foods low in carbohydrates. A reporter was also appointed to consider the subject of ash analysis. At the sug- gestion of the committee on organization of a national chemical society, after conference with a similar committee of the American Association for the Advancement of Science, appointed by Section C, a committee of five from this association was appointed, consisting of Messrs. H. W. Wiley, W. C. Stubbs, G. C. Caldwell, Wm. Frear, and H. H. Nicholson, to cooperate with a similar committee of Section C of the American Association for the Advancement of Science, the Washington Chemical 248 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. ITI, No. 2 Society, and similar bodies, in calling a conference of chemists at some central point during the winter to consider the best means of organizing such a society. A committee on ways and means for securing a more thorough chemical analysis of foods and feeding stuffs, and also on a chemical exhibit at the World’s Columbian Exposition was appointed at this meeting. Fifty-two members and visitors were present. We have now given in some detail an account of the circumstances which gave rise to the establishment of this association, and have followed its development for the first seven years of its existence. Someone has said that the first seven years of a child’s life are the most impression- able, and that the impressions received during this period of its life persist to maturity and even to old age. This association was most fortunate, and rapidly grew strong and vigorous under the tutelage of such men as S. W. Johnson, G. C. Caldwell, N. T. Lupton, R. C. Kedzie, E. B. Vorhees, W. O. Atwater, M. A. Scovell, and John A. Myers, to mention only a few of those leaders in chemical work and thought who have ‘“‘passed over the river’, and not to speak of those who have been active members of the association almost from the beginning. Nor should we neglect to mention the fostering care of the various Secre- taries and Assistant Secretaries of the Department of Agriculture. No wonder the baby grew rapidly! Perhaps I might also compare this association to a great invention. Like all really great inventions, its mother was Necessity, and the inventor, as so often happens, had a very hard time and spent several years trying to find good, substantial, enthusiastic and capable promoters. With the exception of six meetings (the first at Philadelphia, 1884; tenth at Chicago, 1893; sixteenth at San Francisco, 1899; twenty-first at St. Louis, 1904; twenty-fourth at Jamestown, 1907; and the twenty- sixth at Denver, 1909,) all of the conventions of the association have been held at Washington. For many reasons, which may occur to all of you, the habit of meeting at Washington has contributed in no small measure to the effectiveness of the work which the association has accomplished. It is noteworthy that at least one of the meetings held away from “home”’, namely, the twenty-first, appears from the proceed- ings to have been very well attended and to have been in all respects an excellent meeting. (It is also worthy of mention, perhaps, that this meeting was not marred, nor adorned by any presidential address, President Jaffa not having been present.) Beginning with an average attendance, members and visitors, of less than forty, there were registered as in attendance at the last meeting, the thirty-second, three hundred and twelve, the largest number to date. The average attendance for the ten-year period, seventh to the sixteenth meetings, inclusive, was sixty-four, with a minimum of fifty-two 1919] BRACKETT: PRESIDENTS ADDRESS 249 and a maximum of seventy-seven; and it should be stated that two of these meetings were held away from Washington. The average attend- ance for the next ten-year period, seventeenth to the twenty-sixth meet- ings, inclusive, was one hundred and twenty-two. with a minimum of seventy-four (Denver meeting) and a maximum of two hundred and sey- enteen; and it should be recalled that the St. Louis and Jamestown meet- ings also occurred in this period, each with an attendance of eighty-eight. For the six-year period, twenty-seventh to thirty-second meetings, inclu- sive, the average registration has been two hundred and fifty-six, with a minimum of one hundred and eighty-four and a maximum of three hun- dred and twelve. It would be interesting to plot the curve of attend- ance. For the past three years there has been a steady increase. May we not hope that this increased attendance and correspondingly growing interest in the work of the association will continue? Those who come to the meetings evidently get something worth while, and my experience for the past six years is that not a few of them also give much. Though more than one of my predecessors in this office has no doubt spoken of the incalculable benefits of such gatherings as these, where we have an opportunity, not only to unburden our souls, talk over our difficulties, and help each other in the solution of the numerous problems which con- front us and often become oppressive, but get a chance to learn to know and esteem each other in a friendly way, I cannot refrain from alluding to them again here. Coming as we do from all parts of this great country we should not only, but do, become better Americans as well as better chemists and closer friends. I have no doubt that the annual meetings of this association, by affording opportunities for friendly as well as professional intercourse, have done much to eradicate any delu- sions or foolish prejudices we of one part of the country may have harbored with regard to another, and have thus helped to weld the North, East, South, and West into one common country. Beginning with three committees of three on the three subjects, phos- phoric acid, nitrogen, and potash, there were added at the third meeting two more committees on food stuffs and dairy products, and again at the fourth meeting two more committees on fermented liquors and sugar. At the fifth meeting these seven committees were replaced by reporters. By a division at the seventh meeting of the subject of cattle foods, an eighth reporter was added. At the eleventh meeting a reporter on tannin, the ninth, was appointed, and for the first time associate reporters are mentioned, their duties not being defined, except to fit themselves for reporters. At the fourteenth meeting (1897) the name “‘reporter’’ was superseded by “referee.” At the seventeenth meeting the subjects of liquors and foods were divided so as to necessitate the appointment of about fifteen new associate referees, at the sugges- 250 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 tion of Dr. A. L. Winton. As you all know, the scope of the work has continued to grow steadily, until now we have seventeen referees and forty-one associate referees, handling about sixteen different subjects with their sub-divisions. This surely does not mean that we have yet covered the whole field of the subjects which this association may legitimately cultivate, nor even that we are doing all that needs to be done in connection with the lines of work we are now pursuing. The recent work undertaken by Dr. I. K. Phelps on the study of the Kjeldahl method suggests the possibility, if not, indeed, the necessity, of the study of many other processes which are ill understood. As the revision of our methods of analysis is to come up tomorrow for discussion and final adoption in the form of a report of the committee on editing methods of analysis of the association, and since this is prob- ably the most vital matter to be considered at this meeting, it seems proper and timely to say a few words on this subject. As is well known to many of you, the methods of analysis were, up to the twelfth meet- ing, published as a part of the proceedings, when they were, largely on account of the increased volume of the proceedings, ordered to be printed separately, and appeared in 1895 as Bulletin 46, Division of Chem- istry, U. S. Department of Agriculture. Bulletin 46, Revised, was is- sued in 1899. The Provisional Methods of Food Analysis were printed as Bulletin 65, Division of Chemistry, in 1903. The methods of the association, complete, were first published as Bulletin 107, Division of Chemistry in 1907, and the revised edition in 1908, and finally a reprint issued in 1912. It has now, therefore, been eight years since our methods, as a whole, have been printed in revised form. I am sure that we and all other chemists in the country are looking forward with great satis- faction to the issue of a complete revision to date. It is without possi- bility of question that the attendance on the meeting when our commit- tee makes its final report should be as large and representative as we can make it. There should be full, but brief and clear-cut, discussion, coupled with constructive criticism of our committee’s report. In discussing the citrate and oxalate methods for “‘reverted’’ phosphoric acid at the third meeting of agricultural chemists at Cincinnati in 1881, Professor N. W. Lord said he considered that a great part of the dis- crepancies and variations was due, not so much to the solvent, but to the differences in carrying out the details. He hoped that whatever method was adopted, the details would be so clearly and plainly stated it would be impossible to have any more trouble on this account. Mr. N. T. Lupton, in his presidential address at the ninth meeting of this association, 1892, says: “First, a word in reference to conformity to our prescribed methods. All agree that these should be strictly adhered to in our published 1919] BRACKETT: PRESIDENT’S ADDRESS 251 results. As public analysts, our certificates are taken in evidence by the courts of the country, and their money value is of very great im- portance to manufacturers and others, to say nothing of their higher value as reliable expressions of the composition of the great variety of substances submitted to us for analysis. If these methods are unsatis- factory to any member of the association, the amplest opportunity is given in our annual meetings for the presentation of objections and the adoption of changes, should such be desirable in the opinion of this body. Until such changes are officially made, it is the duty of each and every analyst to conform strictly to the instructions given in our bulletins. Should it, for good and sufficient reasons, be deemed advis- able to deviate from official methods, let the results be so stated in unequivocal language. Those who have had much experience in such matters know the disposition of young chemists to make short cuts, and, in a spirit of perfect honesty, to make slight changes here and there in the methods prescribed for their guidance. “A close examination of the published work of the association will show, I think, a gratifying approximation to uniformity in results, and yet there are perplexing discrepancies not easily accounted for, otherwise than on the ground of deviation from the directions given or want of care in the work.” These quotations would seem to indicate definitely one phase of the discussion of our committee’s work. I may add further: Would it, or would it not, be the part of wisdom, and tend to greater uniformity of results, to limit the number of methods designated as official to one (of course giving two as official where we have both a good volumetric and a good gravimetric method), and that one admitted by the majority to be the best and most reliable and workable in the hands of the inex- perienced, but skilled and careful analyst, the other methods being understood as for use only as checks? Would it, or would it not, be conducive to uniformity of results to allow less latitude in the use of different reagents in some of our best methods, for example the Kjeldahl for nitrogen? The serious and careful consideration of our committee’s report certainly deserves our sanest and best thought. The respect for and pride in the work of this association which I hope to arouse in the younger members, the increased appreciation of the value and far reaching results of the work it has done and is doing, and the inspiration to give the best they can offer of their talents to its future development and enlarged usefulness, would fall short of my desires should I fail to mention some of the more important movements in which this association has had a share and has often been the deter- mining factor. Among these movements were the formation of a national chemical society—the American Chemical Society; the establishment of 252 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 food and drug standards; the passage of the food and drugs act; the adoption of uniform volumetric standards; the passage of the bill estab- lishing the Bureau of Standards; the adoption of uniform methods for fertilizer control, and for reporting analytical results; investigation of methods for testing chemical reagents; the adoption of methods for sampling fertilizers, etc. What of the future? Is there nothing left to be done? Shall we now “Wrap the drapery of our couches about us and lie down to pleasant dreams’? At the sixth meeting of this association, Mr. John A. Myers, in his presidential address, said, “We strive, first, to secure accuracy; second, rapidity and adaptability of methods of analysis; and, thirdly, uniformity in stating results.” It seems to me there is much left to do along all three of these lines. In addition, some of our methods are more expensive than they need be. We need a new method for potash, with platinum at its present price. Sodium compounds for potassium would seem to be called for with potash selling for such prices as now prevail. The following suggestion from Dr. Wm. Frear, in his presi- dential address at the fourteenth meeting, is well worthy of restatement and careful consideration: ‘‘Primarily, the work of the association has been chiefly along the lines of importance to the official chemist. This must still be, to a large extent, true of the association’s work. But it will fail of its high opportunities and choose an ideal lower than it may properly select if its work be not pushed also, in a large measure, along more distinctly scientific lines.” This should now have even a greater appeal than formerly, since we have our own Journal. Dr. L. L. Van Slyke, in his presidential address at the eighteenth meeting, gave some suggestions and recommendations, which are as appropriate at this time as when they were made: ‘‘Some of our methods of analysis must remain unimproved, unless individual members of the association devote much time to special research.” The field is still open and the opportunities numerous. Again, “I believe,” he says, “that the time is near at hand when the association should make some effort to encourage and provide for the presentation and discussion at our meetings of papers pertaining to lines of investigations, largely chemical, but not relating necessarily to methods of analysis.” Has not this time now actually arrived? This field has certainly not been overworked yet. In reference to expediting the proceedings of the asso- ciation, there are one or two of Dr. Van Slyke’s suggestions which we would do well to follow: ‘What we want presented on the floor of our convention are sharp, — clear, comprehensive statements of results, with only such detailed expla- nations as are needed to make those statements readily understood.” Again, “If time may not be saved, interest at least will be increased 1919] BRACKETT: PRESIDENT’S ADDRESS 253 if referees and all speakers will develop enough lung power to let others know what they are saying. “One other suggestion in connection with saving time is the matter of promptness in attendance at the opening of every session.”’ I wish heartily to endorse all of these suggestions made by Dr. Van Slyke. In closing these remarks as president, I should fall short of my duty, as painful as it is, did I not announce to you thus officially the passing of three of our active members since we met last in regular annual ses- sion as an association: Professor Robert James Davidson, of the Virginia Polytechnic Insti- tute, Blacksburg, Va., ‘“‘closed his earthly career suddenly, December 19, 1915, leaving a beautiful and beneficent memory”’!. After attendance upon the eighth meeting, 1891, Professor Davidson became an active member of the association, and rarely missed a meeting. He served as president at the twentieth meeting, and was otherwise prominent in the work of the association. Dr. George Edward Patrick, chief, dairy laboratory, Bureau of Chem- istry, Washington, D. C., who for many years took an active part in the work of this association, passed to his reward March 22, 1916. Mr. Thomas Cuthbert Trescot, of the Bureau of Chemistry, Wash- ington, D. C., whose name appears as a member of the association at its third annual meeting, attended its meetings regularly, and took an active and interested part in collaborative work on nitrogen. After a lingering illness, Mr. Trescot was relieved of all suffering April 21, 1916. All three of these men, our friends and coworkers, were in attendance upon the last meeting of the association. They were faithful to the end, and have received the “Well done, good and faithful servant’. I am sure that you will rise and do honor to them as I call their names again: Robert James Davidson, George Edward Patrick, Thomas Cuth- bert Trescot. 1F. K. Cameron. Science, 1916, 43: 418. FIRST DAY. MONDAY—MORNING SESSION. The thirty-third annual convention of the Association of Official Agri- cultural Chemists was called to order by the President, R. N. Brackett, Clemson College, S. C., on the morning of November 20, 1916, at 10.00 o'clock, at the New Willard, Washington, D. C. REPORT ON DAIRY PRODUCTS". By Harry Kiueter (Dairy and Food Commission, Madison, Wis.), Referee. Owing to the late appointment of the referee, only preliminary work was conducted along the following lines: (1) The refraction of the sour serum of milk, studying particularly the effect, if any, of pasteurization upon the serum. (2) Whether or not natural souring would go on in all cases in a pasteurized milk, especially if the milk was not produced under the most sanitary conditions. * (3) Study of pasteurized milks to which viscogen (sucrate of lime) had been added. RECOMMENDATION. It is recommended that the above studies be continued. REPORT ON FOODS AND FEEDING STUFFS. By A. C. Summers (Department of Agriculture, Commerce and Indus- tries, Columbia, S. C.), Referee. The referee made no report except to recommend that the work be continued for another year with special attention to methods for the determination of crude fiber and crude fat. J. B. Rather? (Agricultural Experiment Station, Fayetteville, Ark.) presented a paper on “The Inosite-Phosphoric Acids of Cottonseed Meal”’®. ! Presented by P. F. Trowbridge. 2 Present address, Standard Oil Company, Chemical Laboratory, Brooklyn, N. Y. * Ark. Agr. Exp. Sta. Bull. 131, 1-20; J. Am. Chem. Soc., 1917, 39: 777-89. 254 1919| CUTLER: FEED ADULTERATION 255 Victor Birckner (Bureau of Chemistry, Washington, D. C.) presented papers on “The Acidimetric Titration of Grain Extracts and Amino Acids in the Presence of Alcohol’! and “‘A Simple Method for Measuring the Acidity of Cereal Products. Its Application to Sulphured and Un- sulphured Oats’’?. REPORT ON FEED ADULTERATION?. By CarLeton Cutter! (Agricultural Experiment Station, W. La Fay- ette, Ind.), Associate Referee. The recommendation of last year ““That the size of sample of scratch and poultry feeds necessary to get concordant results on quantitative grit determination be further investigated” is the basis of the work reported. The results given in the following table were secured on samples sub- sampled ‘by careful quartering, the alternate quarters being discarded until a sample approximating the desired weight was obtained, namely, approximately 10, 25, 50 and 125 grams. Five representative scratch feed samples were chosen and determinations of grit made. Determination of grit in scratch feeds. Se 108 Grams | 25@ Grams | 508 Grams | 1258 GRAMS per cent per cent per cent per cent Sn AR o i atts RA sel slayiea pate siege yen « 6.0 5.0 3.7 3.4 MGT ye War AG Riss alse lais ce fo cke late apcl auape-a sees 11.9 8.5 5.0 4.9 fine 9 Ge tay RRO Ct eRe eS Rec e 4.0 3.5 3.4 3.3 Sd6 5.0.6.0 ROCCO TO DCE Ono nee: 7.4 6.0 5.5 5.0 De a2 co DEO UOD ROD OOO DE CIEE Eire aaa 7.5 5.7 5.0 5.0 PSY CEARG Un teicerat alhacshaveliccictalo beste sisi 7.4 5.7 4.5 4.3 ® Approximately. Samples of 10 and 25 grams, respectively, gave results from 0.6 to 6.9 per cent, and from 0.1 to 3.5 per cent, higher than where 50 grams were used, with respective averages of 2.9 and 1.2 per cent higher. On the 125 gram basis the results were only from 0.1 to 0.5 per cent lower than on the 50 gram samples, averaging 0.2 per cent lower. The work of last year showed that 10 grams of feed was too small an amount to secure concordant results in scratch feeds. This year’s 1 J. Biol. Chem., 1919, 38: 245-54. 2 J. Agr. Research, 1919, 18: 33-49. 3 Presented by R. B. Deemer. ‘Present address, R. F. D. No. 1, Springfield, Vt. 256 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2_ work further confirms this and indicates that not less than 50 grams should be used; increasing the quantity to 125 grams apparently does not increase the accuracy of the determination. RECOMMENDATIONS. It is reeommended— (1) That cooperative samples be sent out during the coming year for the determination of grit and weed seeds in scratch feeds. (2) That a key or outline for the qualitative detection of adulterants in feeding stuffs be prepared and submitted at the next meeting of the association. REPORT ON CRUDE FIBER’. By C. K. Francis? (Agricultural Experiment Station, Stillwater, Okla.), — Associate Referee. The following instructions were sent to collaborators: INSTRUCTIONS TO COLLABORATORS. Material. Two alundum crucibles, two pieces of linen, two samples. DETERMINATION OF CRUDE FIBER. Method I.—Official’. Make the first filtration through the linen furnished and the second through asbestos Specific directions for drying and incinerating are given under Method II. Method II. (One filtration through alundum crucibles.) Add to the fat-free residue in an 800 ce. lipless beaker 200 cc. of boiling 1.25% s phuric acid and boil for 30 minutes, using as condensers round-bottomed flasks filled with cold water or any other form of reflux condenser. A gentle blast of air may be used to overcome foaming. At the end of 30 minutes add 200 ce. of boiling 3.52% sodium hydroxid and continue the boiling for another 30 minutes. Filter rapidly through an’ alundum crucible by the aid of suction. Much time is saved if the clear liquid is decanted after settling for about 5 minutes. Do not permit the solution to become cold. Wash with boiling water, then with a 1.25% solution of hydrochloric acid (14 ec. of hydrochloric acid made up to 500 cc.), until the washings are acid and then wash free from chlorids with hot water. Dry the crucible and contents at 105°-110°C. and weigh. Burn the material until a white or light gray ash is obtained, cool and weigh. The loss in weight is considered to be crude fiber. ) ' Presented by P. F. Trowbridge. 2? Present address, Transcontinental Oil Co., Benedum-Trees Building, Pittsburgh, Pa. 2 J. Assoc. Official Agr. Chemists, 1916, 2: No. 2 (I), 118. 1919] FRANCIS: CRUDE FIBER Results of collaborative work on crude fiber determinations. SAMPLE NO. 1 (COTTONSEED MEAL) bo X ~] SAMPLE NO. 2 (STANDARD WHEAT SHORTS) Texas. ANALYST METHOD I METHOD Ii METHOD 1 METHOD It (OFFICIAL) (PROPOSED) (OFFICIAL) (PROPOSED) per cent per cent per cent per cent F. T. Anderson and L. D. Elliott, Food and Drug Inspection Station, U. S. 10.39 Unable to 4.79 5.20 Appraiser’s Stores, New York, N. Y. filter PRGA erase nay cise ocr cic cl atereh hee ys 10:39 ay eee 4.79 5.20 F. C. Atkinson, American Hominy 10.58 11.25 4.59 4.50 Company, Indianapolis, Ind. 10.55 11.54 4.54 4.38 "eran yE eM gis Taig Seen oe eee 10.56 11.40 4.56 4.44 E. H. Berry, U.S. Food and Drug In- spection Station, 1625 Transporta- 10.60 12.03 4.63 5.08 tion Building, Chicago, Ill. 10.55 12.48 4.58 4.95 AV ETARO Mo ayst- cline, = ise earauae oeiels as 10.57 12.25 4.60 5.01 F. C. Collier. Inland Revenue Depart- 12.20 12.59 4.92 4.80 ment, Ottawa, Canada. PAWETARE seisraie, = 5.0 cts tapes tiarh isos a 12.20* 12.59* 4.92 4.80 F. L. Elliott, Food and Drug Inspec- 11.78 13.34 5.30 6.65 tion Station, U. S. Custom House, 11.70 13.32 5.20 6.49 New Orleans, La. 11.81 13.08 5.19 6.22 12.22 12.86 5.40 6.55 PAV CTAB Stee, fais sofa taiccne spss 355 4 ai0 11.88* 13.15* iets 6.48* J. W. Enochs, Agricultural and Me- PEE Ne eee 4.16 dasd chanical College, College Station, 9.92 12.41 4.54 5.24 oo eletal fie decor 4.66 5.44 PAVELARG cotaree rs Yes 5 Tee ayad Acie oes 9.84 12.41 4.45 5.34 10.43 11.77 4.69 5.17 D. T. Evans, jr., Fort Worth Labora- 10.24 12.35 4.66 5.00 tories, Fort Worth, Texas. LO4 Oar emer ee 4.72 ere PIV ELAR C icra ays « c/oicic ofeye cher « ofe = wicys 10.39 12.06 4.69 5.08 Cornelia Kennedy, University of Min- 10.16 12.48 4.60 5.83 nesota, St. Paul, Minn. 10.50 12.18 4.59 5.55 ENSUITE 0 oa i ie fee ec SN 10.33 12.33 4.60 5.69 LOG eee 4.43 BEES LODZ ese: 4.56 Boe H. G. Lewis, Agricultural College, 10.57 11.75 4.63 4.78 Agricultural College, Miss. IY: Dect ania 4.48 chon 10.27 11.35 4.57 4.71 PRVCLAR Oe! s yes, c Hastie a srajsiela et 10.40 11.55 4.53 4.74 258 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 — Resulls of collaborative work on crude fiber determinalions.—Continued. ANALYST D. G. Morgan, Agricultural Experi- ment Station, Stillwater, Okla. AS CTAR OS © Ohya R. A. Thuma, University of Minne- sota, St. Paul, Minn. AVEFARO nrc shake eas « tn en ae W. E. Thrun, University of Missouri, Columbia, Mo. AVOFORG 55,055» x ew eyeires ts teenies SAMPLE NO. | (COTTONSEED MEAL) METHOD I (OFFICIAL) per cent 10.37 10.25 10.31 METHOD It (PROPOSED) per cent 11.85 11.87 SAMPLE NO. 2 (STANDARD WHEAT SHORTS) METHOD I METHOD I (OFFICIAL) | (PROPOSED) per cent per cent 4.61 4.78 4.62 4.80 4.61 4.79 4.21 5.65 so70 5.47 4.21 5.56 4.80 4.80 4.18 4.20 4.18 5.55 4.64 1919] FRANCIS: CRUDE FIBER 259 Results of collaborative work on crude fiber determinations.—Continued. SAMPLE NO. 1 SAMPLE NO. 2 (STANDARD (COTTONSEED MEAL) WHEAT SHORTS) ANALYST METHOD I METHOD II METHOD I METHOD 1 } (OFFICIAL) (PROPOSED) (OFFICIAL) (PROPOSED) ! per cent per cent per cent per cent G. P. Walton, Bureau of Chemistry, 11.03 11.59 4.93 5.18 Washington, D. C. 11.11 12.27 4.81 5.52 LUI 0 ASB eee oe Pan wae 11.07 11.93 4.87 5.35 W. E. Weber, Department of Agricul- 10.22 11.12 4.81 5.30 ture, Harrisburg, Pa. 10.32 11.76 4.76 5.18 PRMERARE Saco (ep aie aint te Seki el 10.27 11.44 4.78 5.24 C. A. Wells, Experiment Station, Ex- 11.25 12.00 4.80 4.66 periment, Ga. 11.42 11.94 4.88 4.75 LRWETRD ES Spee tat Ee RS Soe 11.33 11.97 4.84 4.70 Average of individual determinations. . 10.48 11.86 4.64 SETl Average excluding determinations SEC) gr ea a 10.46 11.75 4.58 5.04 COMMENT BY COLLABORATORS. E. H. Berry: Method I apparently gives more concordant results. It is impossible to say whether the erratic results with Method II are due to the method of boiling or to the use of the alundum crucible. F. L. Elliott: Considerable difficulty was experienced in filtering through the alundum crucible, especially with Sample 1. With Sample 2 the filtration by this method was fairly satisfactory. L. D. Elliott: So much difficulty was experienced with the filtration with Method II that the results are unreliable. J. W. Enochs: Much trouble was experienced in filtering with Method II, especially in the case of the cottonseed meal; in several tests with the latter the liquid could not be filtered at all. J. W. Kellogg: Method II was followed very carefully, using distilled water. With Method I the second filtration was made through linen filters instead of asbestos. In both methods an electric muffle was used for the incineration. In Method I a 35 cc. porcelain Gooch crucible was used. The filtration through the alundum crucibles was very unsatisfactory because of the time required. H.G. Lewis: No difficulty was experienced in filtering through the alundum crucible. J. M. Pickel: Because of the slowness of filtration and the consequent imperfect washing, the determination by Method II is without value, except as indicating the difficulties of the method when applied to cottonseed meal. F. B. Porter, Forti Worth Laboratories: 1 agree with Mr. Evans’ statement that the alundum crucible is impractical because of the long time required for filtration. G. G. Remsburg: No difficulty in filtering was experienced with Sample 2 in either method. 260 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. ITI, No. 2 Sample 1 filtered with little difficulty by the official method when care was exercised that the crucible did not filter free of solution at any time. With the one filtration method this sample filtered with difficulty after one-half the solution had passed through. Frequent washing with 1.25% hydrochloric acid was necessary to complete the filtration. J. H. Roop: Erlenmeyer flasks (800 ce.), fitted with reflux condensers, were used instead of beakers. Method II was more rapid than Method I and gave higher results. In Sample 1 only 83.50% would pass through a 1 mm. sieve. This sample was insufli- ciently ground to permit concordant results on duplicates, although care was taken to secure as uniform a sample as possible. J.W. Sample: In the first attempts with Method II, great difficulty was experienced in filtering but later by allowing the solution to stand for 10 minutes more rapid filtra- tion was secured, the filtration and washing being completed in 1 hour. The solutions were not allowed to cool to any great extent before filtering. In each case the first washing was made with 250 ce. of dilute hydrochloric acid, made according to the directions. The material was then washed thoroughly with hot water, dried at 105° for 6 hours, weighed and heated for 1 hour, and weighed again. On material of this character the new method seems to give much higher results than the official method, and as the filtration is so much slower the new method is not considered good. R.W. Thalcher: Both of our analysts secured higher results by Method IT than by Method I, which is undoubtedly due to the methods of digestion rather than to the filtration, as one of our analysts by reversing the methods of filtration obtained results which checked those obtained by following the methods outlined. W. E. Thrun: Filtering through alundum crucibles is very slow unless the particles are allowed to settle for at least 5 minutes; after washing with the hydrochloric acid — solution filtration takes place more rapidly. Linen was used for both filtrations in Method I. It is thought that the mechanical condition of the cottonseed meal may account for the wide results. G. P. Walton: There was little apparent difference in the speed of filtration between the first and last times these crucibles were used. A solution containing 1.25 grams of sulphuric acid per 100 ce. is not a 1.25% solution, nor is a solution containing 3.52 grams of sodium hydroxid per 100 cc. a 3.52% solution. It is apparent, therefore, that if the theory of the neutralization method is to be made conformable to the official method, using 1.25% solution, a weaker solution than 3.52% sodium hydroxid must be used, viz., one with a normality coefficient of 0.8896, corresponding to an approxi- mately 3.42% solution. No reason is apparent for using a 1.25% hydrochloric acid solution for the final washing out of those substances soluble in hot 1.25% sulphuric acid solution. A solution of hydrochloric acid with a normality coefficient of 0.2566, corresponding in acidity to a 1.25% sulphuric acid solution, would contain approxi- mately 0.93% of hydrochloric acid. RECOMMENDATIONS. It is reeommended— (1) That the one filtration method be further investigated. (2) That the subject of a uniform filtering medium be further studied. | 1919) BROWNE: REPORT ON SUGAR 261 REPORT ON SUGAR By C. A. Browne (New York Sugar Trade Laboratory, New York, N. Y.), Referee. Experiments were continued upon the modification of the Clerget method whereby 50 cc. of the solution are inverted in a 50 to 55 ce. flask with 5 cc. of concentrated hydrochloric acid and the volume com- pleted to 55 cc. after inversion. The results indicate the accuracy of this modification and a cooperative test should be made by the next referee. The referee made a detailed study of methods for determining small quantities of reducing sugars in the presence of large amounts of sucrose. Under this condition sucrose undergoes a slight hydrolysis or decomposi- tion, producing substances which reduce the alkaline copper reagent. The amount of this hydrolysis depends on the amount of reducing sugar present. Existing methods for this determination (Meissl, Hiller, Munson and Walker, and others) generally require a preliminary assay in order to determine the concentration necessary for the final deter- mination. The referee is attempting to work out a method that will hold for all concentrations. Tt has been shown by the referee? that the amount of copper reduced by sucrose was directly proportional to the weight of sucrose and inversely proportional to the weight of reducing sugar in the solution taken for analysis. This reducing action was found for cane molasses, sirup, and similar products not to be expressed exactly by the quantity & but by the modification aon in which S represents the milligrams of sucrose by Clerget, G the uncorrected milligrams of glucose corres- ponding to the weight of reduced copper, and a an analytical constant which is unchanged for any given method. If Allihn’s method is used the value of a is 40. Curves plotted for a wide range of values using the formula a5 begin to deviate from actual results obtained by analysis when the sucrose approaches very large amounts and reducing sugars very smal] amounts. Concentrations of sucrose up to 9 grams in 25 cc. show an increase in reducing action; between 9 and 15 grams the reducing action is approximately constant; and with amounts exceeding 15 grams in 25 cc. the action undergoes a decrease. This decrease may be due to the formation of complex sucrates of copper and potassium, whose coefficient of dissociation decreases as the sucrose content increases as suggested by Maquenne®, but in the referee’s opinion it is due to the 1 Presented by W. D. Horne. 2 J. Am. Chem. Soc., 1906, 28: 450. 3 Compt. rend., 1915, 161: 617-23. 262 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [ Vol. IIT, No. 2 excess of sucrose holding a part of the reduced cuprous oxid in a state of colloidal suspension. While it is impossible to establish any simple numerical relationships between the reducing power of sucrose and invert sugar for all concen- trations, the referee has found it possible to do this algebraically for as high concentrations as are necessary in ordinary analysis. If it is desired to correct for the retarding influence of high concentra- tions of sucrose upon the reduction when using Allihn’s method, the formula ar is modified to ¢ 45 + The quantity iat= is negligible TABLE 1. Correction for the reducing action of sucrose on Allihn’s copper solution. CORRECTION (C) TAKEN s GLUCOSE ———_—_- CORRECTED FouND (G) 3S? | etucosr (G—C) ERROR G+40 sucROSE (5S) GLUCOSE 1000 Gt mgs mgs. mgs. mgs. mgs. mgs. 250 50.0 52.3 2.7 49.6 — 0.4 250 100.0 102.8 1.8 101.0 + 1.0 250 150.0 151.8 1.3 150.5 + 0.5 250 200.0 199.0 1.0 198.0 — 2.0 500 100.0 104.5 3.2 101.3 + 1.3 500 150.0 153.2 2.6 150.6 + 0.6 500 200.0 203.2 2.0 201.2 + 1.2 500 250.0 251.3 1.6 249.7 — 0.3 1000 50.0 60.3 9.9 50.4 + 0.4 1000 100.0 108.2 6.7 101.5 +15 — 1000 200.0 205.3 4.1 201.2 + 1.2 1000 250.0 252.0 3.4 248.6 —14 2000 50.0 66.6 18.3 48.3 —17 2000 100.0 113.7 12.9 100.8 + 0.8 2000 200.0 207.5 8.1 199.4 — 0.6 2000 250.0 255.5 6.7 248.8 —1.2 3000 00.0 26.5 28.6 — 2.1 —2,1 3000 50.5 75.8 24.9 50.9 + 0.4 3000 202.2 212.8 11.8 201.0 —1.2 5938 62.5 101.0 39.2 61.8 —0.7 5625 49.9 91.6 41.6 50.0 + 0.1 5938 25.6 70.3 45.0 25.3 — 0.3 6000 00.0 41.7 41.8 —0.1 —01 6125 11.2 58.2 46.5 ut yg + 0.5 6125 25.0 69.7 46.8 22.9 —21 9000 00.0 50.2 48.2 2.0 + 2.0 12000 00.0 49.0 44.6 4.4 + 4.4 15000 00.0 50.9 42.7 8.2 + 8.2 1919] BROWNE: REPORT ON SUGAR 263 for amounts of sucrose less than 1 gram, but with quantities much above 1 gram retardation in its reducing power becomes so pronounced that a correction must be made. The application of this formula to the analysis of known mixtures of sucrose and dextrose is given in Table 1. The referee has investigated a method of determining reduced copper! by igniting the crucible containing the cuprous oxid and then plunging the hot crucible into vapors of methyl alcohol. Reduction to metallic copper is almost instantaneous. A similar method using ethyl alcohol has been proposed by Wedderburn?. The alcohol used for reduction should be changed frequently since oxidation products may interfere with complete reduction to metallic copper. There is also danger of decomposition of alcohol and deposition of carbon if the crucible is too hot or the alcohol too strong. The method is not one which can be depended upon in the hands of unskilled chemists. Its simplicity renders it serviceable and it is recommended to the next referee for further study. RECOMMENDATIONS. It is reeommended— (1) That the modifications proposed last year for determining sucrose by acid and invertase inversion be further studied. (2) That the work upon determining small amounts of reducing sugars in the presence of sucrose be continued. (3) That the methods of determining copper by reduction of the oxid in alcohol vapors be investigated. (4) That the optical methods for estimating raffinose in beet products be examined with special reference to hydrolysis by means of enzyms. RECOMMENDATIONS ON SUGAR. By W. D. Horne (National Sugar Refining Company, Yonkers, N. Y.). It is recommended— (1) That the referee on sugar investigate the mixing of raw sugar samples in a mortar instead of on a plate, to diminish moisture changes. (2) That the referee on sugar investigate the defecation with the mini- mum amount of lead subacetate requisite to cause flocculation, and to avoid an excessive quantity for producing a lighter colored filtrate than is necessary to obtain a reliable reading. 1Z. Zuckerind. Bohmen, 1897-8, 22: 216-21. 2 J. Ind. Eng. Chem., 1915, 7: 610-1. 264 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No.2 — (3) That polarizations be checked by readings above and below, rather than by averaging. (4) That polarizations be made at 20°C. or that temperature correc- tion for levulose be included with that for sucrose. (5) That the Bureau of Standards be asked to certify to the most advisable Baumé scale, and that it be adopted. WATER IN FOODS AND FEEDING STUFFS. By W. J. McGee (Bureau of Chemistry, Food and Drug Inspection Station, San Juan, P. R.), Associate Referee. The referee found it impossible to undertake any work upon this subject. For future work, it is recommended that a study be made of the best method of determining the moisture in each kind of food or feeding stuff. The determination of moisture in cottonseed meal has been under consideration by the Cottonseed Crushers’ Association for some time. It might be profitable to make a study of that. Even though it may take a long time to accomplish it, the considera- tion of the official method for determining moisture as compared with newer methods which might be developed for each kind of food or feed- ing stuff might be valuable. INORGANIC PHOSPHORUS IN ANIMAL TISSUE!. By E. B. Forbes (Agricultural Experiment Station, Wooster, Ohio), Referee on Organic and Inorganic Phosphorus in Foods, Feeding Stuffs and Drugs. Further study has been given the magnesia mixture method for inor- ganic phosphorus in animal tissue. The methods of extraction and precipitation used with flesh, blood and brain were those reported in — 19142 with the exception of modifications hereinafter noted. The analyses which form the basis for this report were made by F. M. Beegle (Agricultural Experiment Station, Wooster, Ohio), who worked under — the direction of the referee, and Byron McClelland (Bureau of Chemistry, Washington, D. C.). The first program of work was as indicated by the siete schedule: Prese ented by H. C, Lythge 27. Assoc. Official Agr. Ghamiate, 1916, 1: No. 4 (1), 562-80. 1919] FORBES: PHOSPHORUS IN ANIMAL TISSUE 265 SCHEDULE OF ESTIMATIONS ON WATER-SOLUBLE INORGANIG PHOS- PHORUS IN BLOOD AND BRAIN. Use 10 cc. or Macnesta Mixture; ALLow To StTanp 2 Days: Extract of samples as weighed Extract of samples as weighed. plus phosphate solution". A-1 B-1 A-2 B-2 A-3 B-3 User 50 cc. or Macnesta Mixture; ALLow To STAND 3 Days: Extract of samples as weighed Extract of samples as weighed. plus phosphate solution. C-1 : 1-2 D-2 C-3 D-3 Tables 1 and 2 exhibit complete recovery of added phosphorus in blood with 10 cc. of magnesia mixture and two days standing. If 50 cc. of magnesia mixture are used and allowed to stand three days, the results TABLE 1. Inorganic phosphorus determinations in calf blood. (Analyses by F. M. Beegle.) WEIGHT OF PHOS- PHOSPHORUS ; MAGNESIUM PHORUS RECOVERED SAMPLE | WEIGHT | pyRopHos- PHOS- ADDED AS TREATMENT NUMBER OF PHATE PHORUS | MAGNESIUM] yagNESIUM SAMELE) |! MINUS PYROPHOS- c PYROPHOS- | peR CENT BLANK PHATE Paice grams gram per cent gram gram 10 cc. of magnesia A-1 38.5 O:00687 : | OLOO49 2b | irre es eer See eel eee ee mixture; stood 2 es 39.8 O'007E) [OO05S25) aces... | ieee ese 3 days. 35.5 OL0063F '|)'O!002945 |) S55: ... || Been Sires. Average wo || ecto O\QO5OGH ette-roes |e eee. 10 cc. of magnesia B-1 37.2 OOH Sdooe 0.0448 | 0.0445 | ..... mixture; stood 2 B-2 35.3 O0510)5| arc 0.0448 | 0.0448 | ..... days. B-3 41.9 0:0528) || teres 0.0448 | 0.0454 | ..... IAVELAZOW| ees t | ioaee el eames 0.0448 | 0.0449 | 100.22 50 cc. of magnesia C-1 9:0 ||. O:0089 It OL00G 25s) al ereye aoe eects mixture; stood 3 C-2 47.9 O!OLOTA | ROO0G 225 Ear me | tts een pee days. Ces iaiile 34.3) «|| (0:0084:5|JOIOOBRP I acs. Sh buble ce omarpel earache Average||" |) 2. 54 ODOC 44H Ts ee ON a ee 50 cc. of magnesia D-1 41.3 0.0536 | ..... 0.0448 | 0.0440 | ..... mixture; stood 3 D-2 30.3 QO51 20 eva. 0.0448 OOZES He et days. D-3 42.5 O05465)\\ ec 0.0448 O044 Te ere. AVETAREN sey iMacs 3s ell ee aa 0.0448 | 0.0443 | 98.88 Include also blanks on the reagents and checks on the phosphate solution added. 266 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. ITI, No. 2 are higher. Inorganic phosphorus apparently is split off, but complete recovery of added phosphorus is noted. Table 3 indicates that the precipitation of inorganic phosphorus increases with the amount of magnesia mixture added and the length of time allowed for precipitation, probably through cleavage of organic compounds. TABLE 2. Inorganic phosphorus determinations in blood. (Analyses by Byron McClelland.) WEIGHT OF PHOS- RROO TEESE SAMPLE || WEIGHT Pietreoe" INORGANIC ome TREATMENT NUMBER Joke ae Eeaae esac MAGNESIUM BLANK PHATE PHATE. |p 7° “Sea grams gram. per cent gram gram 10 ce. of magnesia A-1 30:00)")' 10:0068 7] O:00585") ee nesia mixture; C-2 10.3188) 0:0203: ‘| '0:0548° i]. Sori ie ee ea stood 3 days. C-3 8.7014 | 0.0166 0.0532} .:... 4 «.. So Average’ |)! sage] tana 0.0566) 2... 5°] 2 a 50 cc. of mag- D-1 7.9749 | 0.0614 | ..... 0.0448 | 0.0452 | ..... nesia mixture; D-2 S'G1S7 |) D0656° [ven ; 0.0448 | 0.0461 | ..... stood 3 days. D-3 10.6250 | 0.0680 | ..... 0.0448 | 0.0465 | ..... Average!) ois oe Obey lees 0.0448 | 0.0459 1919) FORBES: PHOSPHORUS IN ANIMAL TISSUE 269 The amount of inorganic phosphorus precipitable from hot-water- ammonium-sulphate extracts of blood by the use of magnesia mixture and ammonia increases in proportion to the amount of magnesia mixture used and the length of time allowed for precipitation. The extract is clearly unstable, under these conditions, and the magnesia mixture method of inorganic phosphorus estimation is not applicable to blood. If 50 cc. of magnesia mixture were used, the amount of inorganic phos- phorus precipitable from hot-water-ammonium-sulphate extracts of brain, by the use of magnesia mixture and ammonia, was shown not to vary when two, three or four days time was allowed for precipitation. Two days time was shown to be sufficient for complete precipitation. Added phosphate was completely recovered, whether 10 cc. or 50 ce. of magnesia mixture were used in the precipitation, but the amount of inorganic phosphorus found in the sample was much greater when 50 cc. of magnesia mixture were used. TABLE 6. Inorganic phosphorus delerminalions in brain. (Analyses by Byron McClelland.) WEIGHT OF PHOS- EHOSEHGRUS r MAGNESIUM| PHORUS RECOVERED SAMPLE | WEIGHT | pyropHos- | NORGANIC) appep as TREATMENT NUMBER oF PHATE PHOS- _ | MAGNESIUM] 5+, GNESTUM SAMPLE z PHORUS = 5. | MAGNES MINUS PYROPHOS. PYROPHOS- | pep CENT BLANK PHATE cere grams gram per cent gram gram 10 cc. of magnesia A-1 LOGO O!0200 TI) OObaiaterascs. |) cisece ose mixture; stood 2 A-2 HOON} O!O198: ||| OO5SU ees. s ck 4) See days. eae a ei00o) NOOrg7. | 054s ik ee Average) cae |! bs c0s- 6s O\ODSZMeeae) WP ec erecta 10 cc. of magnesia B-1 10:00) |) 0:0726) |) 7.254. 0.0528 | 0.0528 | ..... mixture; stood 2 B-2 40:00) | (0:0729) ||... 0.0528 | 0.0531 | ..... days. B-3 10;00) | 10.0723 | 22 a5- 0.0528 | 0.0525 | ..... ANETARO scree | sieves 5 |) bs areas 0.0528 | 0.0528 | 100.00 50 ce. of magnesia C-1 OOO SIAC 0222 | OOGISitace es. || ove .s | Santee mixture; stood 3 C-2 1TOOOR 0/0224" | OWS 2eaiese 2 sk. a ates days. C-3 TOOR O0227a | OUGdzr sce o | eels. oe IAVEPARO I skcsc [hk boweus > co OOO 2 iter eno ee eet eee 50 cc. of magnesia | D-1 LUOO) | O01 4 Fear: 0.0494 | 0.0490 | ..... mixture; stood 3 D-2 1O00) | MOLOTL7 *|| tae 0.0494 O:0493 |e 2.. days. D-3 1000s | (OO7LE || See ae 0.0494 | 0.0487 | ..... AVELARE ers areatiaal| Uke cone ay ll earever on 0.0494 | 0.0490 | 99.19 270 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 TABLE 7. Inorganic phosphorus in hot-water-ammonium-sulphate extraci of steer brain. (Analyses by F. M. Beegle.) WEIGHT OF MAGNESIUM SAMPLE WEIGHT | pynopHos- | 1NOBGANIC TREATMENT NUMBER OF PHATE ROSS SAMPLE MINUS PHORUS BLANK grams gram per cent 10.6511 | 0.0109 | 0.0285 10.9059 | 0.0137 | 0.0350 11.9099 | 0.0179 | 0.0419 10 cc. of magnesia mixture and 25 cc. of am- 9.7376 | 0.0136 | 0.0389 10 cc. of magnesia mixture and 25 cc. of am- 1 2 3 4 monium hydrate; stood 3 days before 5 10.4890 | 0.0161 | 0.0428 6 7 8 9 monium hydrate; stood 2 days before filtering. filtering. 10.9132 | 0.0148 | 0.0378 13.0308 | 0.0204 | 0.0436 10.5708 | 0.0175 | 0.0462 11.4914 | 0.0194 | 0.0471 10 cc. of magnesia mixture and 25 cc. of am- monium hydrate; stood 4 days before filtering. 507ce. of magesia mixture: Stood 2) days’: cicccctae ie iv ww phlei tees= 10 10.3200 | 0.0198 | 0.0535 Stood) Sidayss a: o-.scaotn cee e = 11 8.9327 | 0.0177 | 0.0552 Stood’ 4idaysivea:cu:. ee Sees 12 9.4823 | 0.0187 | 0.0549 TABLE 8. Inorganic phosphorus in hot-water-ammonium-sulphate in ertract of brain. (Analyses by F. M. Beegle.) PHOSPHORUS WEIGRT OF RECOVERED MAGNESIUM - SAMPLE | WEIGHT OF | PYROPHOS- pets TREATMENT > - NUMBER SAMPLE PHATE MAGNESIUM MINUS PHORUS PYROPHOS- | peR cE BLANK PHATE 10 cc. of magnesia mix- A-1 : } ture; stood overnight. A-2 11.3821 | 0.0223 | 0.0546 | ..... A-3 Average) (co scenal by aey ae 0:0584 |) Samer 25 cc. of phosphate solu- B-1 10.7052 | 0.0655 | ..... 0.0450 tion added to extract B-2 11.2060 | 0.0693 | ..... 0.0478 before precipitation. B-3 10.1177 | 0.0622 | ..... 0.0428 Average’ [cme al) & apeteb ll eateee: 0.0452 25 cc. of phosphate solu- 1-1 9.7845 | 0.0653 | ..... 0.0466 tion added to sample C-% 11.4200 | 0.0658 | ..... 0.0439 before extraction. C-3 12.1604 | 0.0706 | ..... 0.0473 AV Grages|\\\sicisicie' Petes ree || sere 1919) FORBES: PHOSPHORUS IN ANIMAL TISSUE 271 In the light of present information, then, we must consider that a definite fraction of the organic phosphorus of brain is hydrolyzed in hot- water-ammonium-sulphate extracts by the addition of 50 cc. of mag- nesia mixture and 25 cc. of ammonium hydrate. The magnesia mixture method for inorganic phosphorus estimation, therefore, can not be ap- proved for use on brain without further study and modification. It seems possible that the use of graduated amounts of magnesia mixture in the precipitation of extracts of brain might demonstrate that, within certain ranges of variation in the amount of this reagent used, complete recovery of added phosphorus might be demonstrated without there being variations in the amounts of inorganic phosphorus precipitated from the sample. TABLE 9. Inorganic phosphorus in cold water extract of flesh. (Analyses by F. M. Beegle.) WEIGHT OF MAGNESIUM SAMPLE WEIGHT PYROPHOS- INORGANIC TREATMENT NUMBER oF PHATE PHOS- SAMPLE MINUS PHORUS BLANK grams gram per cent Extract precipitated with 10 cc. of magnesia 1 10.1577 | 0.0571 | 0.1566 mixture and 25 cc. of ammonium hydrate; 2 8.8276 | 0.0500 | 0.1578 stood 2 days before filtering. 3 10.8487 | 0.0609 | 0.1564 BN CEARE! | oie de ches, ill eiayauks 0.1569 Extract precipitated with 10 cc of magnesia 4 12.1040 | 0.0679 | 0.1563 mixture and 25 cc. of ammonium hydrate; 5 9.3100 | 0.0628 | 0.1879 stood 3 days before filtering. 6 9.1685 | 0.0514 | 0.1562 Average (SandiGy | a Pons shee ess 0.1563 Extract precipitated with 10 cc. of magnesia 7 10.3800 | 0.0584 | 0.1567 mixture and 25 cc. of ammonium hydrate; 8 11.5615 | 0.0649 | 0.1563 stood 4 days before filtering. 9 10.6850 | 0.0592 | 0.1543 Average (Ggandi8) ies. eel ee 0.1565 Precipitated with 50 cc. of magnesia mixture: Vara a ERY ed Misi > is ie eee eine oo 10 12.4419 | 0.0707 | 0.1583 SEOOd GAR VBislem fata fo okies sits sty ene 11 12.2478 | 0.0695 | 0.1581 Stood (Sidayast is ssc pete heh |: eee 12 12.0138 | 0.0681 | 0.1579 PSU CEAR CN ere ce w|i celica 0.1581 272 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 2 TABLE 10. Inorganic phosphorus determinations on lean meat in pork chops. (Analyses by Byron McClelland.) WEIGHT | ADDED OF PHOS- PHOS- SAMPLE | WEIGHT | MAGNESIUM| INORGANIC | PHORUS FHORUA TREATMENT NUMBER OF PYROPHOS- PHOS- ADDED /AS ||| RECOVERED SAMPLE PHATE pHoRUS | MAGNESIUM 48 Seeares PY ROPHOS- | MAGNESIUM wie PHATE | PYROPHOS- PHATE grams gram per cent gram per cent No phosphate added; stood A-1 10 0.0419 O:V167, « |) ec cle 3 days before filtration. A-2 10 OLOZ24) (OAT3ON NS: Se eee Phosphate added; stood 3 | B-1 10 ORO yo er 0.0658 99.54 days before filtration. B-2 10 0.1088 | ..... 0.0658 | 101.21 No phosphate added; stood | C-1 10 010442 , |, 01231 55h. ieee 2 days before filtration. C-2 10 010444" ||’ (O!1236'>]| See Phosphate added; stood 2 | D-1 10 OUO84 io eram 0.0658 97.42 days before filtration. D-2 10 O090) |) ese 0.0658 98.33 TABLE 11. Inorganic phosphorus in the cold water extract of flesh. (Analyses by F. M. Beegle.) WEIGHT OF PHOSPHORUS MAGNESIUM RECOVERED SAMPLE | WEIGHT OF | PYROPHOs- | 1NORGANIC TREATMENT NUMBER SAMPLE PHATE EHOS- MAGNESIUM! MINUS PHORUS | py ROPHOS- ae Meare. | PER CENT grams gram per cent gram All samples precipitated A-1 10:2200' |) “O:0531' |) (0:1447 |) eee with 10 cc. of magnesia A-2 11.8576: 0:0618 |) (014529) eee mixture and allowed to’| A-3 14/2849) ,|,0:074:1.s)), 0.0446\2) 4.5.05 ee stand overnight. a Average i) +:% ankle 0.1448. |. .:: dave Ape 25 cc. of phosphate solu- B-1 14.2843 | 0.1184 | ..... 0.0442") eee tion added to extract B-2 1587) OOo 4) ee. 0:0449' |! Siren before precipitation. B-3 13:3000 i OUTOSs |e cttue 0:0436'0i\ Sacer Average:|) .}..uabere| ls wieysetunliee terre 0.0442 | 94.647 25 cc. of phosphate solu- C-1 TZ S650 UO NUL ee oee 0:0437 WNwudttc wie tion added to sample C-2 13.8850 | 0.1161 | ..... 0:0439 luaaee os before extraction. C-3 UI 2S 0b UO LOUS calle ances 0.0428: | co eer AVONAZO| #0 5.20 | Mie os Flere ee 0.0435 | 93.148 1919) FORBES: PHOSPHORUS IN ANIMAL TISSUE 273 RECOM MENDATIONS. It is recommended— (1) That the magnesia mixture method for the estimation of water- soluble inorganic phosphorus in flesh be adopted as an official method, with one minor change of detail, in the interest of economy of reagents, namely, that the amount of magnesia mixture used in extracts from 10 to 12 gram samples be reduced from 50 cc. to 10 cc. (2) That further work be done with the magnesia mixture method on brain. (3) That further work with blood be conducted, not on the whole blood, but on the plasma. No report was made by the referee on the separation of nitrogenous substances. REPORT ON THE SEPARATION OF NITROGENOUS SUB- STANCES IN MILK AND CHEESE. By Leroy S. Patmer! (Agricultural Experiment Station, Columbia, Mo.), Associate Referee. Work has been begun along three lines. (1) As stated last year, 8 to 10 per cent of the total true protein of fresh milk is not recovered either as casein or as heat-coagulable proteins by the methods which this association recognizes. The residual proteins may be recovered, how- ever, by addition of Almen’s tannic acid solution to the filtrate obtained from the heat-coagulable proteins. The origin and character of these residual proteins, as present in fresh milk, should be studied, particu- larly in view of the fact that it is this portion of the milk which increases to such a great extent as protein decomposition progresses in old milk. (2) Preliminary studies indicate that the addition of 0.3 cc. of 10 per cent acetic acid to the neutral casein filtrate, as recommended by the provisional method for albumin, may not be enough to recover the maximum amount of heat-coagulable proteins. (3) Another question being studied is the character and proportion of the heat-coagulable protein which recent investigation? has found to be in colloidal solution in the milk like the casein, or which, at any rate, is retained with the casein when milk is filtered through the Pasteur- Chamberland filter. Such studies are necessary before methods of analysis can be recom- mended for cooperative study by the association. 1Present address, University of Minnesota, University Farm, St. Paul, Minn. *Van Slyke and Bosworth. N. Y. Agr. Exp. Sta. Tech. Bull. 39, (1914), 7 274 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 RECOMMENDATION. It is recommended— (1) That studies be continued leading to the adoption of methods for the determination of the non-casein proteins and the products of protein decomposition in milk. ORIGIN OF THE NEUTRALIZATION PRECIPITATE OF COWS’ MILK. By Leroy S. Pater’ (Agricultural Experiment Station, Columbia, Mo.). The author has shown? that the so-called neutralization precipitate obtained on neutralizing the filtrate obtained from the acetic acid precipitation of the casein from cows’ milk was composed almost entirely of di- and tricalcium phosphates. Recent work by Van Slyke and Bos- worth: offers a very satisfactory explanation of the origin of this precipi- tate, and at the same time clears up several other features of the work which were in doubt at the time the first paper was prepared. The work of these investigators shows that the di- and tricalcium phosphates found in the neutralization precipitate are the result of several chemical reactions between the calcium and phosphorus com- pounds normal to milk and the calcium acetate which is formed as the result of the action of the added acetic acid upon (a) the calcium com- bined with the casein and (b) the calcium chlorid of the milk serum. The sodium hydroxid added to bring about the precipitation of the neutralization precipitate is also involved. The salts entering into the reactions to form the di- and tricalcium phosphates are calcium phos- phate, potassium phosphate and calcium acetate. The presence of mono- magnesium phosphate in milk, as shown by Van Slyke and Bosworth, makes probable the presence of some magnesium phosphate in the neutralization precipitate. The reactions which enter into the formation of the neutralization precipitate may be expressed as follows: (1) CaH,P.0,+CaC,H,.O,=2 CaHPO,+2 C,H,0;. (2) CaC,H,.O,+2 NaOH=Ca(OH).+2 NaC:H;0:. (3) 2 CaHPO,+Ca(OH).=Ca;P,0;+2 H,0. (4) 3 GaC,H,O.+12 KzHPO,+12 NaOH=Ca;P:05+ 6 K;P0,+6 KC,H;0,+4 Na;PO,+12 H.0. As pointed out in my previous paper, casein precipitated by rennet yields a filtrate which does not show the neutralization precipitate. This result also becomes clear in view of studies by Bosworth and Van Slyke. They have found that the paracasein thrown down by rennet Pre sent address, University ‘of Minne sota, “University Farm, St. Paul, Minn. 2 Assoc. Official Agr. Chemists, 1916, 2: No. 1 (1), 4-8. §N. Y. Agr. Exp. Sta. Tech. Bull. 39 (1914), 3-17. ‘Tbid., 37 (1914), 7-10. 1919) TROWBRIDGE: MEAT PRODUCTS 275 contains the calcium in its molecule like the casein thrown down by acid, and also carries down with it the dicalcium phosphate of the colloidal part of the milk. Rennet coagulation leaves no calcium phosphate in the serum, which accounts for the failure of a neutralization precipitate to appear on the addition of sodium hydroxid to the acidified serum. Recent tests by the author have demonstrated, however, that other phosphates in the serum can be detected readily by the addition of calcium acetate and a repetition of the neutralization, when the char- acteristic precipitate appears. The failure of the protein precipitated by rennet to yield a neutraliza- tion precipitate on proper treatment is still unexplained. Paracasein was prepared from skim-milk and thoroughly washed with water until it had a greenish white appearance. Portions were dissolved in acetic acid and ammonium hydroxid, respectively. In the case of the acid solution, a reprecipitation of the protein with sodium hydroxid yielded a filtrate showing the presence of calcium on the addition of ammonium oxalate. This was to be expected from the action of the acetic acid on the calcium combined with the paracasein. There was no evidence of a neutralization precipitate, however, although one would expect that any dicalcium phosphate carried down with the paracasein would be converted into soluble phosphate by the acetic acid and reap- pear on neutralization. Similarly, it would be expected, in the case of the alkaline solution of paracasein, that the dicalcium phosphate held by the paracasein would be converted into tricalcium phosphate; but this should be changed into the soluble phosphate by the acetic acid and reappear as neutralization precipitate on the addition of sodium hydroxid. The only explanation at present available for these results is that the purification of the paracasein was not sufficient to remove enough of the dicalcium phosphate held mechanically to assure its reappearance as a neutralization precipitate. PROGRESS REPORT ON THE SEPARATION OF NITROG- ENOUS SUBSTANCES IN MEAT PRODUCTS. By P. F. Trowsrince! (Agricultural Experiment Station, Columbia, Mo.), Associate Referee. The work is being done by Walter E. Thrun (Agricultural Experiment Station, Columbia, Mo.), the purpose of which is to study the effect of age and condition of the animal upon the composition of the flesh. Ten pounds of flesh are extracted with cold water until the filtrate no longer gives the Biuret reaction. The extract is concentrated by ’ Present address, Agricultural Experiment Station, Agricultural College, N. Dak. 276 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 2 heat on a water bath and the coagulable proteins separated by washing. Three main samples of the flesh are thus obtained: the cold water insolu- ble; the cold water soluble coagulable; and the extract. The cold water insoluble is extracted with alcohol and ether and the alcohol-ether extract examined for lecithin and other nitrogenous compounds. The purine nitrogen! in the residue is determined, and also the amino nitrogen. It is then run for Van Slyke numbers. Tryptophane will also be deter- mined according to the method of Annie Homer?, also tyrosine accord- ing to the method of Plimmer and Eaves*. The aspartic and glu- taminic acids will be determined according to the method of Foreman‘. The coagulable proteins are to be run in the same way as the insoluble residue, except that the purine nitrogen will not have to be deter- mined. The extractives are concentrated to a sirup, but the exact method of handling has not been completely worked out. The samples have been prepared and some of the Van Slyke determinations made, as have many other preliminary determinations and tests. RECOMMENDATION. It is recommended— (1) That these investigations be continued. REPORT ON TESTING CHEMICAL REAGENTS. By C. O. Ew1ne® (Bureau of Chemistry, Washington, D. C.), Referee. The situation with regard to chemical reagents during the past year has been abnormal. Efforts on the part of users and manufacturers should soon clear up difficulties which have been met in obtaining some reagents of a satisfactory quality. The cooperative work of the year consisted in testing the method for the determination of alcohol in pharmaceutical preparations as out- lined in last year’s report. The method will provide for almost all interfering substances except phenol. When it is present the following modification is used®: After the first distillate has been shaken out twice with petroleum ether, draw off the lower alcoholic salt solution into a 200 mil flask, add bromin water to slight excess as shown by a brownish color. Then add a crystal of sodium thiosulphate to remove the excess of bromin and sufficient 50% sodium hydroxid solution to dissolve the precipitated tri-brom-phenol and distil as usual. 1Hall. The Purine Bodies of Food Stuffs. 2nd ed. rev., 1903. 2 J. Biol. Chem., 1915, 22: 369-89. ’ Biochem. J., 1913, 7: 297-310. ‘Tbid., 1914, 8: 463. 5 Present address, United Drug Company, Boston, Mass. 6 J. Ind. Eng. Chem., 1916, 8: 240-1. SK 1 1919) EWING: TESTING CHEMICAL REAGENTS 277 Only two collaborators reported results, and these were not sufficient to justify any positive conclusions as to the general applicability of the method. RECOMMENDATIONS. It is recommended— (1) That work on the determination of alcohol in pharmaceutical preparations be continued. (2) That the method for the determination of the strength of acetic anhydrid as outlined in this year’s report be studied cooperatively. (3) That work on tests of purity for immiscible organic solvents be undertaken. The association adjourned at 12.55 a.m. to reconvene at 1.30 p.m. A. HUGH BRYAN. The death of this well-known food chemist came as a shock to his many friends, the chemical world and the sugar industry at large. His loss is the more keenly felt because death claimed him in the very midst of his life’s task at the age of forty-six years. Mr. Bryan died at his home in Montclair, N. J., on January 19, 1920, after only a brief illness of influ- enza, followed by a complication of acute Bright’s disease and pneumonia. Mr. Bryan was born in Indianapolis on July 27, 1874. After graduating from the public and high schools of his native city he entered Purdue University, from which he received the degrees of B. S. in 1898, A. C. in 1899, and M. S. in 1900. ‘Mr. Bryan’s first position was that of assistant chemist at the Indiana Agricultural Experiment Station, where he remained until 1900, when he became chemist for the American Beet Sugar Company of Colorado. In 1907 he resigned his position with the American Beet Sugar Company to accept an appointment in the Bureau of Chemistry as a sugar chemist. He was made Chief of the Sugar Laboratory of the Bureau of Chemistry in 1909, which position he held until 1913, when he resigned to accept his last position as supervising chemist for Arbuckle Brothers of New York City. There probably was no chemist who was in more intimate touch with the various sides of our American sugar industry than Mr. Bryan. His practical experience brought him into close relations with the beet sugar industry of the West, the cane sugar and cane sirup industry of the South, the sorghum sirup industry of the Central States, the maple sugar and maple sirup industry of the North, and the refining industry of the East. Fully aware of the important role of chemistry in the advancement of all of these branches of sugar chemistry, he was a constant contributor to the chemical literature upon these subjects. His writings comprise a large number of bulletins, circulars and articles upon the subjects of sugar beet, cane, sorghum, maple sirups, honey, etc. He was preparing at the time of his death a book upon the chemistry of coffee, and it is greatly to be regretted that he did not live to complete this work. Mr. Bryan was at numerous times, and as late as 1919, a referee for the Association of Official Agricultural Chemists, and his various reports upon methods of sugar analysis to this association constitute an important part of its published proceedings. Many of his suggestions have been incorporated in the official methods of analysis. In addition to being a member of the Association of Official Agricultural Chemists, Mr. Bryan was a member of the American Chemical Society, the International Commission for Uniform Methods of Sugar Analysis, the In- diana Academy of Sciences and the Washington Academy of Sciences. He was deeply interested in the recent establishment of the Society of Amer- ican Sugar Chemists and Technologists. Only those who have lived in close companionship with Mr. Bryan can speak of the fullness and accuracy of his knowledge, which was always at the service of those who constantly called upon him for advice. The many friends who have enjoyed the hospitality of Mr. and Mrs. Bryan will always remember the happiness of their home life. The sym- pathy of everyone goes out to the faithful wife and son in their bereave- ment. R. E. Doouitr.e. GLUES USED IN AIRPLANE PARTS". Report No. 66, entitled ““Glues Used in Airplane Parts’’, embodies the results of experimental research on the glues used in the manufacture of wooden airplane parts. The experiments were conducted by the Forest Products Laboratory of the United States Forest Service. The report first contains a general statement descriptive of the different kinds of glues used. This statement is followed by detailed descriptions of the following kinds of glues: Animal glues, liquid glues, casein glues, blood albumin glues, and vegetable glues. The detailed discussion in- cludes the description of the manufacture of each type, giving the physi- cal properties, the use of each type of glue, including the proper methods of application, and the proper pressure and temperature required to obtain the best results. Following the detailed discussion of each type is a list of references. A comparison of the different types of glues is next given in tabular form, the table containing the following information: Source, cost, spread, mixing, application, temperature of press, strength, water resistance, staining, and the use of each type in wood-working. The physical tests described in this paper include block shear tests and ply shear tests. Tests were also made to describe the water resistant properties of the glues, keeping quality, odor, jelly strength, viscosity, and other physical properties. A copy of Report No. 66, entitled “Glues Used in Airplane Parts”, may be obtained upon request from the National Advisory Committee for Aeronautics, Washington, D. C. INDUSTRIAL USES OF GLYCEROL. A prominent manufacturer has recently established a fellowship at the — Mellon Institute of Industrial Research, Pittsburgh, Pa., for the purpose of extending the industrial uses of glycerol. It is expected that this investigation will be centered primarily on the use of glycerol to replace alcohol in flavoring extracts. The Mellon Institute is an endowed institution deyoted to scientific research and its application to the industries. Its investigations are con- ducted by means of fellowships, and in such a manner that they are not — affected by any financial or private interest. It is hoped that the results of this glycerol fellowship may be brought to the attention of industrialists — by cooperation with various associations. ' Abs. of Rept. 66, National Advisory Committee for Aeronautics. FIRST DAY. MONDAY—AFTERNOON SESSION. The appointment of the following committees was announced by the president: Auditing committee: J. P. Street, of Connecticut; B. B. Ross, of Ala- bama; and C. B. Lipman, of California. Committee on nominations: C.C. McDonnell, of Washington, D. C.; G.S. Fraps, of Texas; and B. L. Hartwell, of Rhode Island. Committee on resolutions: William Frear, of Pennsylvania; L. L. Van Slyke, of New York; and W. A. Withers, of North Carolina. Committee to invite the Secretary of Agriculture to address the convention: C. H. Jones, of Vermont; P. F. Trowbridge, of Missouri; and W. A. Withers, of North Carolina. REPORT ON PHOSPHORIC ACID. By W. J. Jones, jr. (State Chemist Department, La Fayette, Ind.), Referee, and C. S. Lykrs (Clemson Agricultural College, Clemson College, S. C.), Associate Referee. Details of the proposed plan for the investigation of the determination of reverted phosphoric acid are contained in the report of the associate referee for 19151. This report is one of progress and not for permanent deductions or conclusions. Since the basis of comparison at the present time is a series of results obtained by the use of ammonium citrate solutions, specific gravity 1.09, the first subject for investigation was the study of the methods proposed for preparing neutral ammonium citrate solution, their action and that of proposed substitutes on the phosphoric acid in fertilizers as compared with a solution of 1.09 specific gravity prepared from the neutral triam- monium citrate salt. 1W.J. Jones, jr. J. Assoc. Official Agr. Chemists, 1917, 3: 97. 279 280 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 SOLUTIONS STUDIED. The following solutions, prepared according to the instructions of the various authori- ties, were studied: (1) Official method using corallin as an indicator?. (2) Optional method using alcoholic calcium chlorid and cochineal?. (3) Hand’s method using azolitmin as an indicator’. (4) Solution 1.09 specific gravity from triammonium citrate salt (Baker & Adam- son’s analyzed ammonium citrate, C. P.). (5) Patten’s titration method using phenolphthalein as the indicator in the presence of neutral formaldehyde*. (6) Hildebrand’s method using rosolic acid as the indicator®. (7) Rudnick’s method with N/10 citric acid®. (8) Bosworth’s method with sodium citrate’. PREPARATION OF SOLUTIONS. Ammonium cirate——With the exception of the solution from the triammonium citrate salt, the solutions were prepared from Pfizer’s commercial citric acid and from C. P. ammonia. Curie acid —Analyzed citric acid, C. P., was used and the strength of the solution determined by titrating with N/10 potassium hydroxid, using phenolphthalein as indicator. Sodium citrate-—This solution was prepared by dissolving 305.6086 grams of analyzed sodium citrate, C. P., (this amount being equivalent to the triammonium citrate in a solution of 1.09 specific gravity) and making up to 1 liter. Specific gravity at 20°C. 1.1726. Redistilled rain water was used for all the solutions. METHODS EMPLOYED. The specific gravity was determined by means of a pycnometer. AMMONIA. Twenty-five cc. of the citrate solution at 20°C. were made up to 250 cc., and 25 ce. equivalent to 2.5 ec. of original solution, distilled with 3-5 grams of magnesium oxic into N/2 acid, the excess acid being titrated with N/10 potassium hydroxid, cochineal as an indicator. In Patten’s method ammonia is determined by making 50 cc. up to 250 ce. an distilling 5 ec., equivalent to 1 cc. of original solution, with magnesium oxid. e minations are therefore reported by both methods on this solution. CITRIC ACID. Fifty ce. of the solution were made up to 250 cc. and 5 ce., equivalent to 1 ce. of th original solution, were titrated with N/10 potassium hydroxid in the presence of neutra formaldehyde, using phenolphthalein as an indicator. ‘U.S. Bur. Chem. Bull. 107, rey.: 2 Assoc. Official Agr. re Scihoge, 1916, 4. 2 U.S. Bur. Chem. Cire. 52: ‘1. 4 J. Ind. Eng. Chem., isla” 5: 567. * Ibid., 1914, 6: 577. € Ibid., 1914, 6: 486. 7 Ibid., 1914, 6: 277. 1920] JONES: PHOSPHORIC ACID 281 SOLVENT ACTION OF SOLUTIONS ON PHOSPHORIC ACID IN FERTILIZERS. DESCRIPTION OF SAMPLES. In selecting samples for determining the action of the various solutions on phosphoric acid, 24 samples, of which 22 were on general sale and representative of the products of 8 manufacturers doing an interstate business, were selected. The other 2 samples consisted of tricalcium phosphate and a mixture of the excess inspection samples from the mill room. All samples were ground to pass 100 mesh. TABLE 1. Comparison of various methods of preparation of solution of ammonium citrale. (Analyst, R. B. Deemer, State Chemist Department, La Fayette, Ind.) SPECIFIC GRAMS PER LITER RATIO SOI UTION GRAVITY Tee CA. |) AMMONTIA:TQ Ar 20°C. Ammonia Citric acid CITRIC ACID Jl) Cy oe a 1.09005 43.33 165.82 1-3.826 (2) Optional. Cochineal_________ 1.09007 43.64 166.01 1-3.788 (3) Hand. Azolitmin___________ 1.09009 43.40 168.12 1-3.873 (4) Triammonium citrate salt so- funn eee: eRe AN 8 1.09003 43.73 165.11 1-3.776* Triammonium citrate salt?___.| = ______ = Seay eee 2 1-3.777° (5) Patten. Formaldehyde and phenolphthalein, 1 cc. for BirAtions set 2 2 ot 1.09002 44.08 165.81 1-3.761 USHER TOR TREO) eee eee | 43.75 165.81 1-3.789 (6) Hildebrand. Rosolic acids ____ 1.09004 44.37 165.92 1-3.739 NEUTRALITY Alcoholic | Patten SOLUTION calcium Hand formalde- | Hildebrand Corallin chlorid azolitmin | byde and rosolic and phenol- acid’ cochineal phthalein Orne Tin ee Neutral | Acid Nentrall} 225 == Acid (2) Optional. Cochineal_________ Alkaline | Neutral | Neutral | ______ Acid Eymiiend. Avzolitmin....---..._.| -_._-_- Acid Neutral | ___--- Acid (4) Triammonium citrate salt so- UT Se ae ee eee Alkaline | Neutral | Neutral | _____- Acid Triammonium citrate salt®____ | Alkaline | Neutral | Neutral | ___--_ | ----_- (5) Patten. Formaldehyde and phenolphthalein, 1 cc. for . BREEA IONS fe eee oe Alkaline | Alkaline | Neutral | Neutral Acid ER OLILEAGONS ee sees |) eee ee ere yes (6) Hildebrand. Rosolic acide ___| ______ | _____- ipa ee | | pees es Neutral! ® Theory 1-3.7597. + Twenty-five grams of the salt were made up to 250 ce. and ammonia and citric acid determined as in other hte Solution apres strong odor of ammonia. Moist red litmus paper suspended in calorimeter tube rapidly changed to blue color. 4 True rosolic acid. (The inner anhydrid of <4, 4’, 4”,-tetrahydroxy-3-methyl-triphenylmethane.) 282 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 DETERMINATION OF INSOLUBLE PHOSPHORIC ACID. Digestion —The official method for digesting in citrate solution was followed in detail. Difference due to variation in agitation was eliminated by the use of Huston’s Agitating Machine’, each flask making 2 complete revolutions per minute. SOLUTIONS. Solutions for the determination of total and insoluble phosphoric acid were made by the sulphuric acid, mercuric oxid method and the determinations of phosphoric acid by the gravimetric method. All solutions and determinations were in duplicate with triplicates in case of disagreement. TABLE 2. Descriplion and analyses of samples. POTASH ere TOTAL SAMPLE DESGRIPITON NITROGEN SOLUBLE Peles PHOSPHORIC NUMBER (N IN WATER aoe P.O, (P:0,) (P20s) per cent per cent per cent per cent 3 Completes#@ = _L~ 222): SES 1.0 3.1 3.99 19.29 4 Completes a= ses =.= sapleeee 1.5 2.3 4.20 10.65 5 Complete 2=— 5.2. - ..- See 0.9 2.7 7.39 13.64 7 Gompletete 9+ --- oe 1.6 2.3 1.10 10.14 10 Completes =-2 >= aa 0.9 2.9 4.63 15.94 11 Completes. - =. - 2 ae ee 0.3 3.5 4.52 18.53 14 Completes t a coe a eee 0.9 6.9 1.92 7.26 15 Complete=== ee 1.5 2.0 3.95 9.82 30 Completese = “oe ay 2 eee 1.0 3.2 8.39 13.36 34 Gomplete:= =~ eee 2.6 1.6 6.82 12.13 40 Complete: 23-2 eee 1.0 1.0 1.59 9.57 43 Mixture of inspection samples 1.2 2.4 4.81 10.90 20 Acid phosphate------------- ses aoe 13.91 19.67 — 22 Acid phosphate------~------ =e ae 10.03 19.31 23 Acid phosphate_-_-_-___------- we ats 12.06 19.19 39 Acid phosphate---______~_-- Ses eee 9.26 17.70 45 Acid phosphate, Florida ____- Sas iy 10.68 17.27 41 Precipitated bone________-_- ene J. | elie 38.61 18 (Dankapesteee a. oe) Seana 6.3 2. jon ee 14.16 24 IRE Sh Alo%c) (= ell el 4.1 ==. |) eee 25.70 26 | Ammoniated bone_--_-__-~- 2.6 ia hates = — 21.27 © 33 Raw bonees=scse. 25... se 4.1 Re leeadl | oo == 22.12 k 47 Rock phosphate -________ ~~~ - ae Le Res 31.24 48 Tricalcium phosphate, C. P. - ete sas, || re 41.89 alkali, loses its sensitiveness when added to an ammonium citrate solution, and ealeu- lations based on titrations were found to be incorrect. The use of this method in determining the neutrality of citrate solutions seems to be more or less guess work. 1H. W. Wiley. Principles and Practice of Agricultural Analysis. 2nd ed., 1908, 2: 117. cl J 4 1 1920] JONES: PHOSPHORIC ACID 283 Hand’s azolitmin.—This indicator! did not appear to be sensitive to less than 2 ce. of N/10 ammonia when used in the presence of 5 cc. of ammonium citrate solution. Optional method.—The indicator used in this method is very sensitive, readily reacting to a drop of N/10 alkali or acid. (Referee.—This method, having been in constant use in this department for 25 years, the analysts are undoubtedly more expert in its manipulation.) Patten’s method.—Patten, while stating that neutral formaldehyde is required for use in the titration of citric acid, makes no mention of how this is to be obtained. For- maldehyde which is neutralized in the cold with N/10 alkali shows an acid reaction when used under the conditions imposed by this method—that is, in a solution brought to boiling. : The small aliquot (1 cc.) used in this method necessitates the multiplying of the working error by 1000, in placing results upon a liter basis. Another point that may be mentioned is the fact that in preparing large amounts of citrate (30 liters, as we usually do) to determine accurately the volume and calculate the amount of ammonium hydroxid to add is quite impracticable. Hildebrand’s method.—The solution prepared by this method gave off an appreciable odor of ammonia and quickly turned red litmus paper suspended in the neck of the bottle to blue. Difficulty was experienced in filtering and washing the digestions of tricalcium phosphate with the exception of those in citric acid and sodium citrate. All digestions with sodium citrate of samples containing animal by-products were difficult to filter and wash. Hirsh funnels were used in filtering all insolubles. TENTATIVE CONCLUSIONS. (1) Theoretical composition and ratios are not to be expected from solutions prepared according to official directions when a commercial salt, which may or may not contain impurities, is the basis of the solution. (2) Owing to the small aliquot taken, the method for estimating citric acid greatly multiplies the working error. (3) The results reported are not sufficient to justify permanent deduc- tions but indicate: (a) That the methods studied for the preparation of neutral ammonium citrate do not give a neutral solution, but that two, optional and Patten’s, are slightly acid; two, corallin and Hand’s, appreciably acid. and one, Hildebrand’s, appreciably alkaline when compared with a solution of triammonium citrate salt, specific gravity 1.09. (b) That of the solutions reported, two—optional and Patten’s—are the more nearly neutral ammonium citrate and that in the results re- ported the former is more consistent in its variations. (c) That the ammonia citric acid ratio is not necessarily the controlling factor in results obtained. 1 Prepared according to Cohn. Indicators and Test-Papers. 2nd ed. 1902, 30. 284 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TABLE Description of samples and (Analyst, TRIAMMONIUM CITRATE SALT CORALLIN OPTIONAL HAND PATTEN HILDEBRAND STANDARD 26 26 2 28 25 gal gee) eee hg e lesz] — |s3e| 2 [sez] & =e Pate Zac| & = Bea] & | 3ne| & | saa] & a =| . a A |a ay A |e a per cent per cent| per cent| per cent| per cent| per cent) per cent) per cent) per cent) per cent) per cent 3 9.67 12.43] +2.76| 9.07) —0.60) 11.70}+2.03) 8.21] —1.46} 7.90) —1.77 4 4.13 4.53) +0.40) 3.89} —0.24| 4.65)+0.52) 4.00) —0.13) 3.81) —0.34 5 5.46 5.55) +0.09| 5.26} —0.20} 5.38] —0.08) 5.41) —0.05) 5.12) —0.34 7 6.90 7.10) -+0.20} 6.70} —0.20| 7.22)+0.32) 6.84) —0.06) 6.60} —0.30 10 TAT 7.21; +0.04) 6.97} —0.20| 7.16} —0.01) 1.06) —0.10) 6.92) —0.25 11 6.69 6.72/+0.03| 6.50} —0.19| 6.83)+0.14) 6.77/+0.08) 6.60) —0.09 14 4.17 4.38)+0.21| 3.94) —0.23] 4.45)+0.28) 4.18)+-0.01] 3.81) —0.36 15 3.82 4.22}+0.40| 3.74)—0.08] 4.40/+0.58) 3.93)+0.11] 3.78) —0.05 30 3.81 3.82}+0.01} 3.87/+0.06] 3.73] —0.08) 3.75) —0.06; 3.89}+0.08 34 3.59 3.52} —0.07| 3.55) —0.04; 3.59) + 3.49} —0.10) 3.52) —0.07 40 6.77 Bide == 6.74) —0.03| 6.82}++0.05| 6.72} —0.05) 6.73] —0.04 43 4.03 4.11)+0.08} 3.96} —0.07] 4.04)+0.01) 4.09)+0.06) 3.92) —0.11 20 5.74 5.66] —0.08} 5.57)—0.17| 5.58} —0.17| 5.79) + 5.66) —0.08 22 7.96 7.93) —0.03} 7.89} —0.07| 7.95) —0.01/ 8.07)—0.11} 7.98)-+-0.02 23 5.58 5.63|+0.05|} 5.62}+0.04] 5.76)+0.18} 5.49) —0.09} 5.65} +0.07 39 6.80 6.50} —0.30} 6.72} —0.08] 6.90)+-0.10) 6.67; —0.13) 6.57) —0.23 45 6.45 6.36] —0.09| 6.34|—0.11| 6.39) —0.06) 6.38}—0.07| 6.40) —0.05 41 24.42 24.74! +0.32| 22.33) —2.09| 24.86) +0.44) 22.58) —1.84) 21.42) —3.00 18 9.25 9.14, —0.11] 7.78) —1.47| 9.41/+0.16) 8.06] —1.19} 6.58] —2.67 24 13.19 13.27} -++0.08] 12.10} —1.09} 13.66]-+0.47) 12.41) —0.78) 11.68] —1.51 26 12.38 12.54/-+0.16} 10.21) —2.17} 13.84)+0.46) 10.73) —1.65) 7.59] —4.79 33 12.18 12.45) +0.27| 11.02} —1.16} 12.99/-+-0.81) 11.19) —0.99} 10.23] —1.95 47 1.71 2.11/+0.40} 1.47) —0.24] 1.88}+0.17) 1.41)—0.30] 1.25) —0.46 48 9:82 wend ee ea 8.04) —1.78] 9.47|—0.35) 6.99} —2.83) 6.48) —3.34 Average #2 |eer +0.21}_.___- ay) +0.25)- J. 22)\=0:49|aaee= —0.90 (d) That none of the methods studied indicates correctly the neutral point of ammonium citrate. (e) That continuation of the work, including the results on a larger number of samples, will be necessary before a satisfactory basis for — recommendations will be available. (f) That the results by the citric acid and sodium citrate methods do not with present directions give promise of providing a suitable substi- tute for neutral ammonium citrate in the determination of reverted phosphorie acid. If the results reported are a criterion, the adoption of either the citric 1920) JONES: PHOSPHORIC ACID 285 3. results of digestion with solutions. R. B. Deemer.) poe ee INSOLUBLE PHOSPHORIC ActD (P20s) = = | S ez cs ) mos| & lees] & = = , = sib (ie=aie sh SBS a Aye Agglice Ss) c) i a = Zz A |e percent| per cent| per cenl| per cent) per cent| per cent) per cent) per cent) per cent| per cent| per cent) per cent 5.30] —4.37| 2.73}—6.94| 5.63] 2.87] 6.23} 3.60) 7.09} 7.40) 10.00) 12.57) 3 3.92] —0.21] 2.83} —1.32] 2.32] 1.92) 2.56] 1.80} 2.45} 2.64) 2.53) 3.62) 4 , 4.53) —0.93] 4.03} —1.43] 0.79] 0.70) 0.99) 0.87) 0.84) 1.13) 1.72) 2.22) 5 6.52] —0.38} 5.11)}—1.79} 2.14] 1.94) 2.34) 1.82) 2.20) 2.44) 2.52) 3.93) 7 5.71| —1.46| 5.71/—1.46] 4.14) 4.10) 4.34) 4.15} 4.25) 4.39) 5.60) 5.60) 10 5.29] —1.40) 5.34|—1.35) 7.32) 7.29) 7.51) 7.18) 7.24) 7.41) 8.72) 8.67] 11 3.79] —0.38] 3.07}—1.10) 1.17) 0.96] 1.40) 0.89] 1.16) 1.53) 1.55) 2.27) 14 3.62] —0.20) 2.70) —1.12| 2.05) 1.65) 2.13) 1.47) 1.94) 2.09) 2.25) 3.17) 15 3.28} —0.53] 3.09} —0.72| 1.16) 1.15) 1.10) 1.24) 1.22 1.08) 1.69) 1.88) 3 3.04|—0.55} 2.88}—0.71| 1.72) 1.79] 1.76] 1.72| 1.82) 1.79] 2.27) 2.43) 34 6.13] —0.64) 5.63}—1.14| 1.21) 1.21) 1.24] 1.16) 1.26) 1.25] 1.85) 2.35) 40 3.20] —0.83] 2.82)—1.21} 2.06} 1.98) 2.13} 2.05) 2.00) 2.17] 2.89) 3.27) 43 | | 5.39] —0.35| 4.57|—1.17; 0.02} 0.10) 0.19} 0.18] 0.02) 0.10) 0.37) 1.19) 20 6.51] —1.45) 5.78) —2.18} 1.32] 1.35) 1.39) 1.33} 1.21) 1.30) 2.77) 3.50) 22 4.94) —0.64) 4.30) —1.28} 1.55} 1.50) 1.54) 1.37) 1.64) 1.48) 2.19) 2.83) 23 5.32] —1.48] 4.64)/—2.16] 1.64, 1.94) 1.72) 1.62) 1.77) 1.87] 3.12) 3.80) 39 5.65] —0.80) 4.88] —1.57| 0.14) 0.23) 0.25) 0.20) 0.21; 0.19) 0.94) 1.71) 45 18.88] —5.54| 19.25) —5.17| 14.19) 13.87] 16.28] 13.75) 16.03); 17.19) 19.73) 19.36} 41 6.17] —3.08} 4.30) —4.95} 4.91) 5.02} 6.38) 4.75) 6.10) 7.58) 7.99] 9.86) 18 8.30] —4.89] 4.40] —8.79| 12.51] 12.43) 13.60) 12.04) 13.29) 14.02) 17.40) 21.30] 24 7.18] —5.20| 3.39] 8.99} 8.89} 8.73) 11.06] 7.43] 10.54) 13.68} 14.09) 17.88] 26 7.27| —4.91| 3.83) —8.35] 9.94) 9.67} 11.10) 9.13] 10.93} 11.88) 14.85) 18.29) 33 2.22)}+-0.51| 1.85)+0.14) 29.53] 29.13) 29.77] 29.36) 29.83) 29.99] 29.02) 29.39] 47 8.74| —1.08} 1.83] —7.99} 32.07|--_-__| 33.85] 32.42) 34.90) 35 41 33.15) 40.06) 48 er =i ee eed EE | ee a es ee acid or sodium citrate methods studied would necessitate an entire revision of the established understanding of reverted phosphoric acid and also of the value placed on this ingredient for crop production. The results reported, coupled with many others obtained in the course of twenty-four years’ work in fertilizer inspection, tend to confirm the personal opinion of the referee that the most satisfactory solution of this problem will be to base the solvent solution not on its neutrality but on a definite amount of ammonia per liter in a solution of ammonium citrate, specific gravity 1.09, the latter determined by weight. 286 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 RECOMMENDATIONS. It is recommended— (1) That the study of the preparation of neutral ammonium citrate solution, its use in determining reverted phosphoric acid and possible substitutes for it in this determination be continued. (2) That in view of the conditions resulting from the European war, whereby the price of molybdic acid has been more than quadrupled and 100 per cent molybdic acid practically removed from the market, the referee study the determination of phosphoric acid with a view to recom- mending an optional method not requiring the use of molybdic acid. REPORT ON PHOSPHORIC ACID IN BASIC SLAG?. By C.S. Lyxns? (Clemson Agricultural College, Clemson College, 8. C.), Associate Referee on Phosphoric Acid. Twelve sets of three different samples of basic slag were prepared and ten sets of samples were sent to chemists who had signified their desire to cooperate in the work. The instructions to collaborators are substantially the same as those sent out by the referee for 1915 with the following addition under (B)?: (d) In a 500 ce. volumetric flask dissolve 2 grams of the slag in about 5 cc. of nitric acid and 20 cc. of sulphuric acid, rotating the flask so that no lumps form and so that each particle will be attacked by acid. Cool and make up to mark. Determine phos- phoric acid by the volumetric method, using an aliquot of 20 ce. TABLE 1. Moisture determinations. ANALYST SAMPLE PER CENT J. O. Clarke, Department of Agriculture, Atlanta, Ga._________-- = ee 3 0.42 R. E. Pennell and C. S. Lykes, Clemson Agricultural College, Glemson ‘Gollege:;SNGis-2 202 _ ee See eee F ee 3 0.53 * Average of 2 determinations in all cases. Particular attention was directed to the preparation and concentration of solutions used in the volumetric analysis. The 500 cc. Wagner flasks must have a neck width of at least 20 mm. and must be marked at least 8 cm. below the mouth. The filtration ’ Presented by P. F. Trowbridge. ? Present address, Solvay Process Company, Syracuse, N. Y. § J. Assoc. Official Agr. Chemists, 1917, 3: 90. 1920) LYKES: PHOSPHORIC ACID IN BASIC SLAG 287 must be performed immediately after rotating 30 minutes. The use is recommended of a folded No. 597, S. & S. filter paper sufficiently large to permit the whole quantity of liquid to be poured on the filter at once. If the beaker containing the mixture of phosphate and molybdate solu- tions is heated on the water bath to 60-70°C., a precipitate free from silicic acid results. If heating is continued for a considerably longer time, the precipitate will often be mixed with silicic acid, especially when the molybdate solution is not added to the filtrate immediately. TABLE 2. Determinaiion of total phosphoric acid. Se : te ame Tor oe) 3 ong ESE sag E: g £28 EeS3| Beo8 x. 4. z= Sa | EASES ANALYST = eax Sai =. Rasa Mg g-5 REQi| BESS oc = =e = ove ome 2 —_ 2~— = _-n2 = —_ 2 _ 2s a] 2 ao = oe) = 2.00 | So & = |) Ss: ses 288 © 8.2 -4 | 8.205 2 ete = Sis 8 Stes ets a| Oo ‘s) ro) a ro) ro) | | per cent | per cent | percent | per cent | per cent | per cent | per cent aa) Clarke <2 5-2. 222 TPG 37 (016:50° | TSAR 1G 40° |) 16:46 |e ee | Ta es 2 WEST R202 || L7SO8) VES | W7UOo|)) ss 2-25) set = Bierce dea Do | Le OVCa eli Osea LG.OLS |= = R. E. Pennell and Ge SAbykes*2-6 2: 1 | 16.904 | 16.43e | _____ PEES2E || £ eet 16.304 | 16.72¢ PAP ny SN a i iC i eS OSE: || = 4S" 16.69 | 17.13¢ Slliesse | LeOre | see 57 Ga IO) | ar 16.804 | 17.13¢ = Assoc. Official Agr. Chemists, Melnods, 1916, 2. b J. Assoc. Official Agr. Chemists, 1917, 3: 90. © Average of 2 determinations. 4 Average of 4 determinations. © Average of 3 determinations. ! Average of 5 determinations. ® Average of 8 determinations. b Average of 7 determinations. TABLE 3. Determination of available phosphoric acid. | METHOD ANALYST SAMPLE | MOLYBDATE : ‘ aC puanel, ic Lorenz Tron citrate | per cent per cent per cent per cent pers Clarke: sree e ees Fok 1 14.76 14.90 14.85 14.72 2 15.07 15.27 15.22 15.28 3 15.42 15.57 15.458 15.40 R. E. Pennell and C. S. Lykes 1 14.82¢ 14.55° 14.908 14.93¢ 2 15.19¢ 14.85° 15.39¢ 15.25¢ 3 15.46¢ 15.16° 15.784 15.56! « J. Assoc. Official Agr. Chemists, 1917, 3: 92. » Average of 2 determinations, unless otherwise indicated. © Average of 4 determinations. 4 Average of 5 determinations. © Average of 6 determinations. ! Average of 3 determinations. 288 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 DISCUSSION. It will be noted that the instructions recommend the use of 5 cc. of sodium acetate solution before precipitating with magnesia mixture in the gravimetric determinations according to method I, 5 (g). Reference to Table 1 shows that all results were high, averaging about 0.3 per cent above those obtained by the volumetric method. In all cases the pre- cipitate was redissolved and the iron determined, the consequent cor- rected results falling about 0.3 per cent below those by the volumetric method. A perusal of the table shows a considerable variation in iron content, the samples which contained a high percentage of iron running the highest. When, however, correction for iron is made, results which are uniform, though low, are obtained. Some difficulty was experienced in getting the slags into solution with hydrochloric and nitric acid, due to precipitation and caking of silica on the bottom of the flask, but by constant shaking this difficulty was overcome. When samples were put into solution by method I, 5 (g), all volumetric determinations were found to yield concordant results, no variation from the average by more than 0.29 per cent being noted. On all volumetric work from nitric-sulphuric acid solutions, results from a greater number of analyses were found to agree even more closely, except on Sample 1. With samples of slag put into solution with sulphuric and nitric acid, the precipitate of phosphomolybdate was formed in larger crystals, and for this reason there was less danger of loss on filtration by running through the filter medium. All filtering was done by suction, and a thin pad of pure white asbestos fiber was used. Due to abnormally high results obtained when filter paper or paper pulp is employed as the filter medium, the associate referee used only asbestos. The concordant results of the several methods and the results by the iron citrate method should be noted. All methods for available phos- phoric acid show very close agreement. The iron citrate method is more rapid, easier of manipulation, and results in extremely close checks. RECOMMENDATIONS. It is reeommended— (1) That the volumetric method from sulphuric and nitric acid solu- tion be adopted as official for total phosphoric acid in basic slag. (2) That the continuation of the work on the availability of phos- phoric acid in basic slag be governed by the report of the field committee on this work. 1920) HASKINS: REPORT ON NITROGEN 289 REPORT ON NITROGEN. By H. D. Hasxrtns (Agricultural Experiment Station, Amherst, Mass.), Referee. The work on nitrogen has been along the lines recommended by the referee for 1915. The work on nitrogen is presented in two reports: First, a study of the Jones and Street methods for organic nitrogen activity, which was conducted under the direction of the referee; second, a study of the ferrous-sulphate-zinc-soda method for nitrates, and the use of sodium sulphate in place of potassium sulphate in the Gunning method and its modifications, which was planned and carried out by the associate referee, Mr. R. B. Deemer. ORGANIC NITROGEN ACTIVITY BY THE JONES AND STREET METHODS. By reference to the proceedings of the association, it is found that the first serious consideration of organic nitrogen availability in mixed fer- tilizers was in 1895, when S. H. T. Hayes submitted results of a study of ten different organic nitrogenous substances. This included the use of alkaline, neutral and acid permanganate solutions, and barium hydrate solution, as well as a fractional treatment with sulphuric acid. Attention was called to the necessity of treating equal amounts of nitrogen in the permanganate digestions. In 1896 J. P. Street gave some comparisons of organic nitrogen avail- ability by the pepsin hydrochloric acid method and by vegetation experi- ments, and as a result of Street and Davidson’s investigation of Hayes’ method, it was recommended that the referee for 1897 be requested to investigate the permanganate methods. In 1897 J. P. Street reported results of determinations carried out on 1 gram of material, which showed an utter lack of agreement. Certain modifications by H. B. Slade gave more promising results, and it was recommended that the method be further studied. In 1898 R. J. Davidson submitted results obtained by the pepsin hydrochloric acid and the permanganate methods. C. H. Jones for the first time presented his modifications, which consisted of the use of 100 ec. of a solution made up by using 16 grams of potassium perman- ganate and 150 grams of sodium hydroxid per liter, employing enough sample to furnish 45 mg. of nitrogen, digesting, below boiling, for 1 hour and allowing 1 hour for distillation. He also submitted his interpreta- tion of results by this method, classing the materials as either good or questionable, according to whether they ran over or under 50 per cent in nitrogen activity. In 1899 F. S. Shiver concluded that the neutral gave more promising 290 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 results than the alkaline permanganate method and recommended a further study of the neutral method. At this time the Jones method employed 45 mg. and the Street method 75 mg. of nitrogen. In 1900 W. R. Perkins called attention to the necessity of washing the fertilizer with water to remove soluble phosphates, etc., before treatment with the permanganate solution. Although results of the year’s work were discouraging, he recommended that both of the permanganate methods be further studied. In 1901 W. R. Perkins studied the two methods and reported results which were not entirely satisfactory, but with some misgivings he recommended that the neutral method be adopted as provisional, and that further study be given the alkaline, or Jones, method. In 1902 F. W. Morse concluded that the alkaline permanganate method was worthy of further investigation, and recommended that both methods be further studied with a view to getting more concordant results. In 1903 F. W. Morse continued the study of the two methods, recommend- ing some modifications of the neutral method, and also further study of both the neutral and the alkaline methods. In 1904 C. H. Jones sent out three samples of fertilizer and received reports of results from 15 chemists. Certain improvements and modi- fications were made in the alkaline method, and it was recommended that this method be adopted by the association as provisional. In 1905 no cooperative work was done on these methods and no report was made to the association. In 1906 J. H. Gibboney studied the two methods with two mixed fertilizers of known make-up, and received the cooperation of fifteen chemists. He stated that, although the alkaline method gives low results, it presents greater possibilities than the neutral procedure, and recom- mended that both be studied further. In 1907, 1908 and 1909 no work was done on either method. In 1910 C. H. Jones revived the interest of the association in the two methods. In the case of the neutral method, the important changes since its adoption as provisional were the employment of an equivalent of 45 mg. of water insoluble organic nitrogen for the determination, and the elimination of the filter paper during the permanganate digestion. The alkaline method had been modified by the use of material equivalent to 50 mg. of water-insoluble organic nitrogen. With mixed goods an increase in the strength of permanganate solution from 1.6 to 2.5 per cent was effected. The data furnished showed a very satisfactory agree- ment between the two revised methods and carefully conducted pot experiments by Messrs. Hartwell and Pemberton on twelve complete fertilizers. In 1911 no additional work was done on the two methods. ao — A a ee 1920) HASKINS: REPORT ON NITROGEN 291 In 1912 C. L. Hare studied the two methods by the use of four fer- tilizer samples, seven analysts taking part in the work. The neutral permanganate method gave fairly uniform results in the hands of different analysts, but the alkaline permanganate methods gave satisfac- tory results only with base goods and mixed fertilizers. In 1913 C. L. Hare was of the opinion that the methods furnished a fair basis for arriving at the relative activity of the organic nitrogen in various fertilizers, and that either method would serve to differentiate the good forms of nitrogen from the bad, but expressed the belief that the small number of results secured by the association would hardly warrant a positive recommendation that the methods be adopted as official. _In 1914 R. N. Brackett reported that the Jones (alkaline) method gave results more in accordance with pot tests than the Street (neutral) method, and that, while the Jones method was shorter, the Street method in the hands of inexperienced workers gave more uniform results. He recommended that both methods be studied in order to increase the accuracy of the determination of the water-insoluble organic nitrogen and to overcome the difficulties of distillation, and further that they be adopted as official. In 1915 R. N. Brackett reported results which made it evident that the two methods in the hands of analysts who were familiar with the manipulation involved and who had equipment suitable for the work served to differentiate the good from the bad nitrogen. Three samples of mixed fertilizer were used in the study of the two methods by your present referee. The make-up of the fertilizers was as follows: Sample 1. Wieipght-ol.cottonseed meall —- "| ___'2_ Sema aes ere 50 per cent Weight of 16 per cent acid phosphate________________- 25 per cent Weight of high grade red dried blood__________________ 25 per cent Sample 2. Weight of ground garbage tankage, degreased__________ 50 per cent Weight of 16 per cent acid phosphate_-________________ 25 per cent Weightof cottonseed mealse0 “eee 25 per cent Sample 3. Weight of high grade red dried blood_________________- 25 per cent Weight of 16 per cent acid phosphate_________________ 50 per cent Weight of concentrated tankage-_-__-__-_-___-__--_---- 25 per cent 292 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 A set of these samples with the following instructions was sent to the collaborators: You are requested to analyze each sample for total nitrogen according to the official Gunning method, the water-insoluble organic nitrogen and the available nitrogen according to the two methods enclosed herewith. Each sample should also be analyzed for its moisture content. In the case of the Street method report results as follows—total nitrogen, ammo- niacal nitrogen, water-insoluble organic nitrogen, permanganate-insoluble nitrogen; in the case of the Jones method—total nitrogen, ammoniacal nitrogen, water-in- soluble organic nitrogen, nitrogen liberated by the alkaline permanganate solution. ALKALINE PERMANGANATE METHOD FOR ORGANIC NITROGEN ACTIVITY. (a) Transfer an amount of material equivalent to 50 mg. of water-insoluble organic nitrogen! to an 11 em. No. 597, 5. &S. filter paper as a preliminary to washing with water. In case of Samples 1 and 2, wash with 40 cc. of ether on dry filter paper, using 10 cc. with each washing. Wash once with alcohol to displace ether, and wash with successive portions of water at room temperature until the filtrate amounts to about 250 cc. (b) Transfer the residue, with 20 cc. of water, by means of a 20 cc. pipette drawn out to a fine point, so that complete delivery takes about 25 seconds, the filter paper being opened and laid in an elliptical shaped piece of tin bent so that the washings will be properly conducted into the mouth of the flask, to a 500-600 cc. Kjeldahl dis- tillation flask (round-bottomed preferred, but if flat-bottomed is used, incline at an angle of 30°). Add 15-20 small glass beads or fragments of pumice stone to prevent bumping, and 100 cc. of alkaline permanganate solution (25 grams of pure potassium permanganate and 150 grams of sodium hydroxid, separately dissolyed in water, the solution cooled, mixed and made to volume of 1 liter). A piece of paraflin about the size of a small pea may be added if danger from frothing is apparent. Connect with an upright condenser to which a receiver containing standard acid has been attached. It is recommended that the distillate be collected in a 100 cc. graduated open cylinder. Digest slowly, below distillation point, with very low flame, using coarse wire gauze and asbestos paper between flask and flame, for at least 30 minutes. Gradually raise the temperature and when danger (if any) from frothing has ceased, distil until 95 ce. of distillate are obtained and titrate as usual. In cases where a tendency to froth is noticed, lengthen the digestion period and no trouble will be experienced when the distillation is begun. During the digestion, gently rotate the flask occasionally, par- ticularly if the material shows a tendency to adhere to the sides. It is recommended that as nearly as possible 90 minutes be taken for the digestion and distillation. The nitrogen thus obtained is the active water-insoluble organic nitrogen. MODIFIED NEUTRAL PERMANGANATE METHOD FOR THE AVAILABILITY OF ORGANIC NITROGEN. Weigh a quantity of the fertilizer equivalent to 50 mg. of water insoluble organic nitrogen! on a moistened 11 cm. No. 597, S. & S. filter paper, and wash with successive portions of water at room temperature until the filtrates amount to 250 ce. Transfer Determined by extracting 1 gram of the material on an 11 em. No. 597, Ss: & Ss. filler aper with water at room temperature, until the filirate amounts to about 250 ce. Determine nitrogen in the residue, making a correction for the nitrogen in the filler paper, if necessary. 1920) HASKINS: Availability of the organic nitrogen of Sample 1. TABLE uF REPORT ON NITROGEN 293 ANALYST E. F. Berger, Agricultural Ex- |. periment Station, E. Lan- AIP MATCH ee eee se E. A. DeWindt, Agricultural Experiment Station, E. eiansmng) Mich. =. 2222. C. H. Jones, Agricultural Ex- | perment Station, Burling- Gea treme es ot ol L. S. Walker, Agricultural Ex- periment Station, Amherst, LV eros 2 Siete eee eae R. B. Deemer, State Chemist Ppvement, La Fayette, rid Se ewe Pe tA et R. E. Ingham, Virginia-Caro- lina Chemical Co., Rich- MONG AVA. = ee fe 2 Ee F. N. Smalley, Southern Cot- ton Oil Co., Savannah, Ga. C. A. Jacobson, Agricultural Experiment Station, Reno, Oto ee ee H. S. Chilton and I. D. Ses- sums, Agricultural and Me- chanical College, Agricul- tural College, Miss.________ J. H. Perry, Agricultural Ex- periment Station, Orono, L. W. Bradley, Department of Agriculture, Atlanta, Ga.___ PA VELAPO + ssaee 58S IE | ei S 5 > xO Bt /debs Wee 2 eee ie Emig} Beliagy 5 E = <0 E 2 2 = per cent!| per cenl per cent) per cent ate || AROS) |) SY (e539) | ees ese = Il (ihyats" 6.72 | 6.52 | 0.07 | 6.20 T2934) 0.07 | 2-2 6:40 5.53 | 6.68 | 0.07 | 6.28 6.01 | 6.85 | 0.05 | 6.51 6.32 | 6.83 | 0.04 | 6.40 ae | ee ee 7 6:69) || 7.13 | ---- | 6:81 Sees | eLO) |) SeaemOSAG CLO -O8 | SeaenGD 5.30 | 6.72 | 0.07 | 6.32 5.97 | 6.85 | 0.06 | 6.40 6.51 6.87 | 0.06 | 6.48 manganate Nitrogen liberated per cent 4.35 4.30 4.09 3.91 4.67 3.85 4.76 |= | Zo S 3 De per cent} per cent) per cent 67.0 | 0.43 | 93.0 65.0 | 0.50 | 92.0 66.0 | 0.43 | 93.0 61.0 | 0.18 | 97.0 75.0 | 0.32 | 95.0 72.0 | 0.55 | 91.0 60.0 | 0.20 | 97.0 74.0 S s 69.0 | 0.47 | 93.0 70.0 | 0.55 | 92.0 51.0 | 0.73 | 89.0 51.0 | 0.29 | 95.0 65.1 | 0.42 | 93.4 * Unable to interpret results. 294 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 the insoluble residue with 25 cc. of tepid water (the bulk of the residue may be removed by a small spatula, care being taken not to scrape or rough up the paper, which may be laid on a piece of tin bent in the form of an ellipse; a 25 cc. pipette that will deliver its volume of water in about 25 seconds will be found satisfactory) to a 300 cc. low- form Griffin beaker, add 1 gram of dry sodium carbonate and 100 cc. of 2% perman- ganate solution. Digest in a steam or hot water bath for 30 minutes at the tempera- ture of boiling water, covering the beaker with a watch glass and setting well down into the bath so that the level of the liquid in the beaker is below that of the bath. Stir twice at intervals of 10 minutes. At the end of the digestion remove from the bath, add 100 cc. of cold water and filter through a heavy 15 or 18.5 cm. folded filter. Wash small quantities at a time with cold water until the total filtrate amounts to about 400 cc. Determine nitrogen in the residue and filter, correcting for the nitrogen of the filter. Results have been received from eleven different sources, including thirteen different analysts. Very little or no comment was made by any of the collaborators. TABLE 2. Availability of the organic nitrogen of Sample 2. - JONES | STREET B ed 19 10 to 2 os £2 ae ANALYST ° E ial =s a6 2s & z no S' o\)| se) esas ae hae: | a MS *2h) 325/728 E | 2 8 | 88 £35 | a5 | £38 S 5 5 ¢ cS) 3 Sic £ 28 z 25 per cent| per cenl| per cent\ per cent) per cenl| per cent| per cent) per cent Ee WS Berens es. ee Le .--- | 2.90 | 0.05 | 2.55 | 1.03 | 40.0} 0.36 | 86.0 | Egy Nel DON Vivre bape wel eee Da 7.09 | ---. | -=-- | 2.68) |, 0.98.|) 37:0)) 80:57) 7o0 Go: Jones: aes ee 5.60 | 2.84 | 0.04 | 2.60 | 1.04 | 40.0 | 0.30 | 88.0 SS: Walkers. -smeere 3 Ss 5.60 | 3.15 | ---. | 2.54 | 0.96 | 38.0 | 0.23 | 90.0 RR? B? Deemer. Jase ees ws 4.42 | 2.93 | 0.11 | 2.60 | 1.10 | 42.0 | 0.43 | 83.0 Bi. Ingham ji4ee 598 4.80 | 2.98 | 0.05 | 2.65 | 1.29 | 49.0 | 0.52 | 81.0 102 IN DAB Ys oe eee a 5.30 | 2.89 | 0.07 | 2.53 » * 0.55 | 78.0 GsA-TacobsanA eisai 3: a2) (2se8e| eb 258 OSB SA * * Swift & Company ANGIVEL AN ss = ae e8 eee 5.35 | 3.18 | ---. | 2.87 | 1.30 | 45.0} 0.39 | 86.0 Analyst “Bess ea oes .... | 3.21 | ..-_ | 2.88] 1.31 ),45.0.) 0.417) 66g Hl. S. Chilton and I. D. Ses- SUB = 22. 253 ee GE See 5.38 | 3.06 | _... | 2.48 | 0.84 | 34.0 | 0.44 | 82.0 Ss PeIY = eee ena eee 5.98 | 3.04 | 0.10 | 2.68 | 0.77 | 29.0 | 0.26 | 90.0 L. W. Bradley__.-.--.----- _.| 4.90 | 3.02 | 0.05 | 2.63 | 0.87 | 33.0 | 0.60 | 77.0 AVETHDO=W ect eeee ee 5.44 | 3.02 | 0.07 | 2.64 | 1.03 | 38.8 | 0.42 | 83.8 * Unable to interpret results. 1920} HASKINS: REPORT ON NITROGEN 295 TABLE 3. Availability of the organic nitrogen of Sample 3. JONES STREET Zz 5 a 8 Sa Bes ane Le ANALYST S g az | 4. £s ES 3 S = 5 z zZ = = 2 S & z 22 z 3 & Zz | So |SEE/S8e| see | 385 = & < = Zz =< Lael < per cent) per cenl| per cent) per cent| per cent| per cent| per cent| per cen O° DB ---- | 6.19 | 0.19 | 3.91 | 3.07 | 79.0 | 0.22 | 94.0 E. A. DeWindt____-_- =a ee! S20) es SOT 2826 72:0) || 'O:16))|, 96:0 Srpeievones:—--.--.--=-+_ == 6.52 | 6.07 | 0.14 | 3.73 | 2.72 | 73.0 | 0.27 | 93.0 2 Wht eee 6.42 | 6.70 | ____ | 3.89 | 2.67 | 69.0 | 0.14 | 96.0 ieis- Deemer. .=--=-=----.-- 5.10 | 6.28 | 0.23 | 3.77 | 3.04 | 81.0 | 0.16 | 96.0 Rei ingham:-—--_/.__.-.___ 5.29 | 6.34 | 0.16 | 3.99 | 3.02 | 76.0 | 0.32 | 92.0 eNesmalley- 22.2222 -....=- 5.50 | 6.31 | 0.20 | 3.72 a a 0.14 | 96.0 epAewacobson- - ----.....-.- 55 | ee |i) A GHD c - Swift & Company Wmdby strane - 26> 20S. 3.4 6.50 | 6.69 | ____ | 4.09 | 3.05 | 75.0 | 0.21 | 95.0 AT RUS SA rr EEE G:63)}) 222) | FOS S01 197 4.0))}) 0:22) | 95:0 H. S. Chilton and I. D. Ses- STL 3S See er 6.00 | 6.59 | ____ | 3.77 | 2.29} 61.0 | 0.54 86.0 EL Dan 6.70 | 6.35 | 0.19 | 4.07 | 2.66 | 65.0 | 0.15 | 96.0 Beaver Bradley__--.-__..__:_- 5.60 | 6.42 | 0.17 | 3.99 | 2.14 | 54.0 | 1.07 | 73.0 LTE i rr 6.18 | 6.41 | 0.18 | 3.91 | 2.78 | 71.2 | 0.30 | 92.3 * Unable to interpret results. Without any attempt to explain the rather wide variation which exists in the total nitrogen and moisture determinations of Sample 1, it may be said that, with two or three exceptions, the total water-insoluble nitrogen and the active water-insoluble nitrogen results by both the methods are reasonably satisfactory. If 50 per cent for the alkaline and 85 per cent for the neutral method be allowed as the lowest figures that would indicate a passing quality of organic nitrogen, on the basis of 50 mg. of water-insoluble nitrogen, both of the methods would class the nitrogen in this sample as good. In the case of Sample 2, although the total and water-insoluble nitrogen may be in as close agreement as in Sample 1, the results by the two methods are not so pleasing. The alkaline method condemns the nitro- gen in all cases, the percentage of activities shown varying from 29 to 296 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. JJI, No. 5 49, with an average of 39 per cent. In the case of the neutral method, six of the determinations would condemn, four might class the nitrogen as suspicious, and two would pass the nitrogen as being derived from a satisfactory source. It would seem that in this particular case, 85 per cent was rather too low an arbitrary figure to indicate nitrogen of passing quality. Apparently this is recognized by the author of the method, Mr. J. P. Street, for he says, “I should view with considerable suspicion one falling appreciably below 90 per cent’. Two analysts obtained barely 90 per cent activity; the others were from 2 to 13 points below this figure on this particular sample. On Sample 3 both methods gave satisfactory results, with the excep- tion of two determinations by the neutral method. TABLE 4. Comparative lests of nitrogen availability. NITROGEN NITROGEN activiry py | ACTIVITY By MATERIAL eosate VEGETATION GANATE eed (oars) 0.42 GRAM METHOD NITROGEN per cent per cent Died blood: (unwashed) 4- 2 =~. 5. see =e 80 pheepiMmanure= = a2 acs oA co aa a ee eo ee 29 toxic Ground tobacco stems2. =p = eee ee ieee eee 18 toxic Dried blood (water-insoluble nitrogen)__________-__---__- 76 75 Gastorpomace= == 2 5.2.22 eons eee ee ee 53 70 Complete fertilizerINo, Sse Se eee es Gompletefertilizer No: \24-2 2.) See Fee ee Pe ee Gompletesfertihizer,Nowd-22= = 5-5 eee ee ee Gomplete:fertilizer:Nou4s_- 2. 2) | eee S25 ee Complete fertihzer-INo.5.-22-. — See eee Complete fertilizersNo- 62-= = 2 ee ee ee Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. Complete fertilizer No. * Mass. Agr. Expt. Sta., Control Ser., Bull. 2: 30. » Certain observed facts indicate that this figure is perhaps too low. 1920) HASKINS: REPORT ON NITROGEN 297 The referee regrets that a lack of time prevented the carrying on of vegetation tests with these three fertilizers. During the past two years, however, the referee has conducted comparative tests which give the relative nitrogen activity by the laboratory methods, as compared with vegetation tests, on the basis of water-insoluble organic nitrogen'. Some tabulated results are submitted taken from these two publica- tions. It will be seen that the neutral method was not included in Table 4, while Table 5 includes both laboratory methods. Many of the sam- ples studied during the two years were selected on account of the sus- picious character of their nitrogen, although other brands were included which were known to contain only nitrogen of good quality. In the case of the vegetation tests the increase in yield of dry matter over the no-nitrogen pots obtained with unwashed dried blood is placed at 80 per cent, and the increases in yield of dry matter due to the other nitrogen sources (derived from the water-insoluble portion of the various fertilizers) are compared with it. A study of Table 4 suggests the following conclusions: (1) In the case of the sheep manure and ground tobacco stems the large amount of organic matter necessary to furnish 0.42 gram of water- insoluble organic nitrogen seemed to have some injurious effect. A much lower yield of dry matter, accompanied by a much reduced root growth, was obtained than on the no-nitrogen pots. (2) All of the brands of complete fertilizer showing a questionable nitrogen activity by the alkaline permanganate method gave a rela- tively low nitrogen activity by the vegetation experiment. the average of sixteen such cases being 46.5 per cent by the alkaline permanganate method, and 32 per cent by the vegetation experiment. It should be noted in this connection that the laboratory method gives full credit in almost every instance. (8) Sample 11 was the only complete fertilizer tested which shows the water-insoluble organic nitrogen to be of good quality when measured by the vegetation experiment, and of suspicious quality by the laboratory method. A large proportion of the organic nitrogen in this brand was derived from cottonseed meal and castor pomace. The results obtained with castor pomace also show that the laboratory method is likely to give results, somewhat too low, on this class of organic ammoniates. This bears out the observations of C. H. Jones of Vermont and B. L. Hart- well of Rhode Island. (4) Allowing 50 per cent as indicating a passing quality of organic nitrogen by the alkaline method, we find that only three out of twenty-five cases (Nos. 10, 6 and 4), which showed nitrogen of poor quality by vegeta- tion experiment, would have failed detection by the alkaline method. 1 Mass. Agr. Expt. Sta., Control Ser., Bull. 2: 28; 4: 32. 298 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TABLE 5. Comparative tests of nitrogen availabilily*. TROMEADE ae MES SE EN | RELATIVE POT NUMBER pate ven earaen NITROGEN | Alkaline Neutral renner | Qvenne, | sorry | peemes: | pera POTS method method gram gram per cent per cent per cent 1 A, B, C, D, E, F, G, H (blood) _| 0.42 0.141 80.00 80.00 bbe ts 5 A, B, C, D (nitrate of soda)____| 0.42 0.146 $2.84 we as 15 A, B (castor pomace)__________ 0.42 0.101 57.31 60.19 pare 60 A, B (cottonseed meal)_________ 0.42 0.121 68.65 50.20 95.70 59 A, B (dried blood, washed) _____ 0.42 0.153 86.80 71.00 pale 62 A, B (garbage tankage)________ 0.42 0.041 23.26 44.81 eS GAB te pk TS ee 0.42 0.029 16.45 48.00 79.30 FO NEE 5 1S POR ae, lh 2 Cen pa tae Te ee! 0.42 0.061 34.61 46.40 82.90 TOMAS. See Pe Py Beek epee 0.42 0.015 8.51 44.00 56.00 DA ABs ee ee ee 0.42 |No increase} ____ 34.00 43.75 1D tA te os ce ee oak ead 0.42 0.105 59.57 54.20 93.40 140A. B=. ote ee so 0.42 0.065 36.88 55.80 82.00 NG JAG BS ieee eee ee seek ee cee 0.42 0.069 39.15 42.40 76.75 PAG Be Sete SPA! _ TEE 0.42 0.061 34.61 54.80 85.00 TONSA TB ties pe ho RS We os 0.42 0.089 50.50 61.00 94.00 DISAB: See eee eee a eee 0.42 0.036 20.43 43.60 ae DQNAS Bea aaS AE UI 2 ds Ee 0.42 0.069 39.15 48.20 80.00 DSTA ee aie ee ee eee 0.42 0.082 46.52 47.20 85.00 D5VAW IB mein Fe SEY 18am: YORE 0.42 |No increase} ____ 37.20 70.00 IGIAG BR eee Sa ae Ue 0.42 0.029 16.45 38.20 62.00 DS IE ie tte ne ae eR hn 0.42 0.055 31.21 40.80 78.00 PAL Al ajetieieesy wilh al asc ty eles eRe LEY 0.42 0.089 50.50 47.80 91.00 SIVACIB Seek alie as Rupee ON Be 0.42 0.098 55.60 52.00 87.00 SIVAN IDE he ae eee See ee 0.42 0.123 69.80 66.00 87.00 SARAUGS Fe Me ee eee ee 0.42 0.019 10.78 47.20 59.00 Py OV Ni LS ts eh a) s ap ph ell en, Sapa 0.42 0.092 52.20 56.00 89.00 (al Neg |b key icin = 2 Telit hae 0.42 0.090 51.06 52.00 88.00 SSRANIB: Soe ee ee ee eee 0.42 0.096 54.47 58.00 95.00 SORA: IB 2. = ae eee eee 5 eee 0.42 0.132 74.89 60.00 95.00 AT ASB OE ee este Deus hi 0.42 0.120 68.09 63.00 88.00 pA Nal 5 aah eee py ie al RA 0.42 0.073 41.42 35.60 91.00 SANA Bae rey meee ae Sere th). 0.42 0.101 57.30 39.00 91.00 45 ASI Poe eee eee eS 0.42 0.110 62.41 39.00 87.00 AG, Ac gia oa, eee = ee ae 0.42 0.094 53.33 50.00 $4.00 48 Alt BeLt 22 eee. eee 0.42 0.097 55.04 56.00 86.00 AQTAN Bo CORON ate tees 4S 0.42 0.125 70.92 62.00 95.00 SLAB 25 eee Basen es YT 0.42 0.037 20.99 26.00 73.00 D2 A, ene cee aeons oe 0.42 0.005 2.84 24.00 67.00 D4 AB. alee 228 0.42 0.024 13.62 44.00 79.00 BHTAL BS: 1a SER oe ee 0.42 0.007 3.97 41.00 50.00 BONA Bcc act eee ee 0.42 0.105 59.57 49.00 89.00 OSUAS Beno aoc ee cet eee mean 0.42 0.083 47.09 37.00 93.00 * Mass. Agr. Expt. Sta., Control Ser., Bull. 4: 36, with the exception of data in last column. » Dried blood (pots 1 A, B, C, D, E, F, G, H), castor pomace (pots 15 A, B), garbage tankage (pots — 62 A, B), received their nitrogen as indicated from unwashed material. © Basis: Nitrogen recovered, dried blood at 80 per cent. 1920) DEEMER: NITROGEN DETERMINATION 299 A study of Table 5 shows that in twenty-five out of forty-one cases (61 per cent) the alkaline permanganate method indicates higher nitro- gen activities than does the vegetation test; that in four out of forty-one cases (9.75 per cent) the alkaline permanganate method fails to dif- ferentiate between the low and high grade forms of organic ammoniates. In three of these cases the fertilizers were known to contain organic vegetable ammoniates (castor pomace and cottonseed meal). It would appear that the chief criticism of the method is that it sometimes gives too low results with organic vegetable ammoniates of good quality. In case of the neutral method, allowing 85 per cent as indicating a passing quality of the organic nitrogen, it is found that out of a total of twenty-one samples containing nitrogen of low grade character, four cases would have failed detection by the neutral method. All the pot work which has been done in the study of these two methods shows that the alkaline gives nitrogen availabilities more closely agreeing with actual vegetation tests. In Table 5 it is found that out of thirty-six tests made, the average nitrogen activity by the vegetation test was 43.07 per cent, by the alka- line method 48.94 per cent, and by the neutral method 81.33 per cent. The results given herewith seem to be in harmony with results obtained at the Rhode Island and Vermont Agricultural Experiment Stations. It seems to the referee that there can be no doubt among the members of the association, who possess properly equipped laboratories for the work and who have given these two methods, as modified and improved to date, a fair trial that both the alkaline and neutral permanganate tests are reliable in differentiating between the good and the poor forms of organic nitrogen. It is the opinion of the referee that efficiency with either method is very largely a question of proper equipment, experience and familiarity with the work, and he does not feel justified in recom- mending further study by the association. He does feel that both methods should be used in control work, particularly on fertilizers con- taining nitrogen of suspicious quality. REPORT ON NITROGEN DETERMINATION. By R. B. Deemer! (State Chemist Department, La Fayette, Ind.), Associate Referee. The work was confined to a study of the ferrous sulphate-zinc-soda method for the determination of nitrogen in sodium nitrate. The ma- terial used was commercial sodium nitrate, ground to pass a 30-mesh sieve. 1 Present address, Bureau of Plant Industry, Washington, D. C. 300 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 INSTRUCTIONS FOR TOTAL NITROGEN DETERMINATION. PREPARATION OF SAMPLE. Pour out the sample and carefully mix, finally spreading thinly and uniformly over glazed paper. In weighing out remove small portions, with point of the spatula, from various parts of the sample thus spread out. MOISTURE. Heat 2 grams of the sample at 125-130°C. to constant weight. NITROGEN. Determine nitrogen by the following methods: Kjeldahl Method Modified. Place 0.5 gram of the sample in a digestion flask, add 30 cc. of sulphuric acid contain- ing 2 grams of salicylic acid, allow to stand at least 30 minutes (overnight if time per- mits) and then add gradually 2 grams of zinc dust, thoroughly shaking the contents of the flask. Digest over a low flame until frothing ceases, then raise the heat to brisk boiling, continue boiling for 10 minutes or until white fumes no longer escape from the neck of the flask. Add 0.7 gram of mercuric oxid and boil briskly for 3 hours; add 10 cc. more of sulphuric acid as required to maintain the yolume above 20 cc. during the digestion. Oxidize with potassium permanganate and complete the distillation as usual. Report time of treatment with salicylic acid mixture. Make determinations in blank on all reagents with above method using 2 grams of sugar; report as cc. of N/2 acid. Ferrous Sulphate-Zine-Soda Method. (a) Place 0.5 gram of the sample in a 600-700 ce. flask, add 200 ce. of water, 5 grams of powdered zinc, 1-2 grams of crystallized ferrous sulphate and 50 cc. of sodium hydroxid solution (36° Baumé). Distil, collect in the usual way in N/10 sulphuric acid and titrate. (b) Repeat (a) placing a plug of glass wool in the neck of the flask before connecting with the distillation apparatus. (c) Weigh out 5 grams of the sample, dissolve in 250 cc. flask, make to volume, pipette out 25 cc. aliquots and proceed as in (a) and (b). (d) In case it is desired to use a small amount of paraffin to prevent frothing, make determinations in addition to (a), (b), and (c), marking them (f), (g), and (h). (e) Repeat (a) without sample and report blank, if any, as ce. N/2 acid. COMMENTS OF COLLABORATORS. W. D. Richardson.—It was our experience in the Kjeldahl modified method to obtain low results. By the ferrous sulphate-zinc-soda method, (a) and (c), we were unable to get satisfactory results. We had trouble with iron coming over in the distillate. Similarly where glass wool was used, (b) and (c), in the neck of the flask, we obtained quite satisfactory results. We do not use connecting bulbs in our distillation apparatus, which may account for some of the trouble experienced. 1920) DEEMER: NITROGEN DETERMINATION 301 TABLE 1. Nitrogen in nitrate of soda. FERROUS SULPHATE-ZINC-SODA ANALYST EIEE DEBE 5 grams 5 grams ere MOPIFIED | 0.5 gram | 0.5 gram | to 250 cc. | to 250 ce no wool and wool —25 ce. —25 ce. no wool and wool H. S. Chilton. Agricultural Col- per cent per cent per cent per cent per cent lege, Agricultural College, Miss. | 15.58" | 16.55" | 16.07 | 16.16 | ___ Swift & Company, Chemical Lab- oratory, Chicago, Ill. PAMIGIVSUS Ace eee ho ee 15.51° ioe 16.39* — 16.56° ANAlYSte by eens = aN 2 eee 15.63° aioe 16.33° ae 16.36° V.B. Hausknecht, Department of Agriculture, Harrisburg, Pa. _-__| 15.65" 16.57 16.02 16.50° 16.40" R. I. Ingham, Virginia-Carolina Chemical Co., Richmond, Va._-_| 15.85 15.98 15.80 15.89 15.83 J. M. Bartlett, Agricultural Ex- periment Station, Orono, Me.___| 16.18 16.24 nae) 16.27 16.20 H. C. Moore, Armour Fertilizer Wiorks: ‘Atlanta, Ga...) ==. 16.15 15.86 16.12 16.01 38 C. H. Jones, Agricultural Experi- ment Station, Burlington, Vt..__ 15.96 15.72 15.72 16.06 15.72 L. B. Johnson, Department of Ag- riculture, Raleigh, N. C.________ 15.69 15.74 15.75 15.69 15.81 L. W. Bradley, Department of Agriculture, Atlanta, Ga.______- 15.88 15.90 15.94 15.83 15.84 E. F. Berger, Agricultural Experi- ment Station, Lansing, Mich..__| 15.52* 15.76 15.74 15.76 Le es E. A. Dewindt, Agricultural Ex- periment Station, Lansing, Mich. 15.63 15.78 15.75 15.72 ee Armour & Company, Chemical Laboratory, Chicago, III. epArEG recent ee eee ee ae © 16.03 16.04 16.03 16.05 OPE Meeker 0! uae 279 Soe ae 16.06 16.05 16.08 16.10 mS rye ee ee BF Ok. & es 16.04 bees 16.05 4s L. S.{Walker, Agricultural Experi- ment Station, Amherst, Mass.___| 16.19 eek ee 15257° 15.59° R. B. Deemer, State Chemist De- partment, La Fayette, Ind. _____ 15.96 15:57" 15.53* 15.75* 15.52* AETHER ES = 852 = ea Aye 15.94 15.92 15.99 15.94 15.94 @ Omitted from average. 302 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TABLE 2. Nitrogen determinations by miscellaneous methods. ANALYST METHOD NITROGEN W. D. Richardson, Swift & Company, per cent Ghicaga, Et - 5 Sah eet” Bey ae ee Es — VY. B. Hausknecht, Department of Agri- culture; Harrishurg, Ba, ee Ulsch-Street +254. 4. 15.94 R. E. Ingham, Virginia-Carolina Chemical Go:, Richmond)iVas_ 22 _ e Gunning modified_____._____ 15.87 C. H. Jones, Agricultural Experiment Sta- tion, Burlington,/Vit@e 2+ 22000. igi 2 ee ae H. C. Moore, Armour Fertilizer Works, Atlanta’ Gate ten ay _ 2) 4 Kjeldahl-Gunning modified___| 16.18 R. B. Deemer, State Chemist Department, La Payette ind: 2. ~ 5. eee ADCO = 22h es eee 15.82 L.S. Walker, Agricultural Experiment Sta- tion;fAmbherst-9)Mass: 2 2 = 2 ok Se ee ZinC-Ir0N — 9g se ee 15.98 Armour & Company, Chemical Labora- tory, Chicago, Ill. a A Green esse ees. ee 15.74 OVE UMeekers.22: 525) 2 ot ee Kjeldahl-Gunning modified___| 15.71 eS 5 Airy eee ot ley ee we L. W. Bradley, Department of Agriculture, Atlanta®:Gazen 4) 5 S-ve#. ope oy Gunning modified___________ 16.00 F. B. Carpenter —It has been our experience, when making nitrogen determinations by the ferrous sulphate method, that if the apparatus is connected up with a plain bulb tube and no glass wool is used the results are invariably high. In this case we used an ordinary trap bulb, and even with this the tendency is higher without glass wool. J. M. Bartlett.—The ferrous sulphate-zinc-soda method seems to be a rapid, accurate and desirable method for this material. I used a 750 cc. flask tipped at an angle of 45° for the distillation and had no trouble with frothing. H. C. Moore-—We had some trouble with the ferrous sulphate-zinc-soda method on account of the sample frothing over. The distillation was finally made at a very low temperature. Paul Rudnick.—The ferrous sulphate-zinc-soda method not only gives consistently uniform results, but works very smoothly and easily. H. D. Haskins.—A number of tests were lost by frothing in cases where glass wool was not used. I believe it is a precaution that is quite necessary. RELIABILITY OF THE METHOD. A comparison of the results (Table 1) obtained by the method under study with those obtained by the Kjeldahl modified and other methods (Table 2) shows that reliable results are obtained. 1920) DEEMER: NITROGEN DETERMINATION 303 TABLE 3. Substitution of sodium sulphate for potassium sulphate. (Nitrogen as per cent.) | FERTILIZERS FEEDS GUNNING MODIFIED GUNNING Potassium Sodium SAMPLE KJELDAHL f R SAMPLE sulphate sulphate NUMBER MODIFIED Potassium Sodium NUMBER plus plus sulphate sulphate 0.2 gram 0.2 gram copper copper sulphate sulphate 1 1.68 1.72 1.73 26 2.28 2.17 2 1.75 1.68 1.69 27 2.89 2.86 3 0.84 0.85 0.88 28 1.47 1.47 4 0.64 0.63 0.64 29 2.59 2.05 5 1.07 1.09 1.08 30 1.61 1.61 6 121 1.14 es lr 31 1.47 1.47 a 1.68 1.58 1.65 32 1.50 1.44 8 1.77 1.72 1.74 33 2.48 2.49 9 0.93 0.88 0.87 34 2.48 2.45 10 0.50 0.46 0.46 35 2.40 2.45 11 Ds 1.79 1.78 36 3.79 3.72 12 0.75 0.67 0.69 37 2.78 2.76 13 1.24 1.23 1.28 38 1.64 1.68 14 1.30 1.25 1.28 39 2.45 2.44 15 0.95 0.89 0.90 40 1.47 1.59 16 1.28 1.28 1.28 41 1.24 1.26 17 1.20 1.20 1.21 42 2.70 2.67 18 1.12 1.11 1.09 43 1.46 1.40 19 1.51 1.49 1.49 44 6.84 6.88 20 po ae 1.23 1.24 45 6.35 6.43 21 0.53 0.53 0.49 46 2.55 2.60 22 1.14 1.12 1.14 47 2.70 2.79 23 0.47 0.51 0.51 24 0.89 0.89 0.90 25 1.72 1.76 1.80 LOSS OF NITROGEN. It is the opinion of the referee, Mr. Haskins, that loss of nitrogen may occur in the Kjeldahl modified method when used upon nitrate of soda if the oxidation is completed with permanganate in the usual way. Mr. L. S. Walker, of his laboratory, reports 16.19%, (2 determinations) without the use of permanganate. Using permanganate I obtained 15.96% (3 determinations); and 16.00% (3 determinations) when no per- manganate was used. SUBSTITUTION OF SODIUM SULPHATE FOR POTASSIUM SULPHATE. A brief study was made of the use of sodium sulphate in place of potassium sulphate in the Gunning method and its modifications. An average of 25 samples of fertilizers with the Kjeldahl modified gave 1.17% uitrogen; with the Gunning modified and potas- sium sulphate 1.11%; and with sodium sulphate 1.16% nitrogen (Table 3). The average 304 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 of 22 feed samples gave with the Gunning method and potassium sulphate 2.77% and with sodium sulphate 2.60% of nitrogen. When sodium sulphate is used some trouble is experienced with caking upon cooling after the digestion; but if the molecular equivalent of potassium sulphate is used, and care taken to dilute as soon as sufficiently cooled, this does not cause great inconvenience. RECOMMENDATIONS. It is reeommended— (1) That the tentative ferrous sulphate-zinc-soda method be adopted as official. Owing to the conflicting results of previous work it is sug- gested that the use of glass wool in the neck of the distillation flask receive further study. (2) That further study be made of the effect of permanganate at the end of the digestion in the Kjeldahl modified method when used on nitrate of soda. (3) That the use of sodium sulphate in the Gunning method in place of potassium sulphate be tried out on a variety of organic substances of known origin and of difficult oxidation. SUBSTITUTION OF SODIUM SULPHATE FOR POTASSIUM SULPHATE IN THE KJELDAHL-GUNNING-ARNOLD METHOD FOR THE DETERMINATION OF AMMONIA IN FERTILIZERS. By T. D. Jarreww! (State College of Agriculture, College Park, Md.). In view of the present high price of potassium sulphate and the com- parative low price of sodium sulphate, tests were made of the use of the latter in the Kjeldahl-Gunning-Arnold method for the determination of ammonia in fertilizers. Latshaw? reports results on ten different samples, substituting sodium sulphate for potassium sulphate, using the Gunning copper method. These results show a very close agreement between the two processes and he concludes that sodium sulphate can be used as well as potassium sulphate as a catalytic agent for raising the boiling point of sulphuric acid in any kind of material. Latshaw discusses the comparative costs of these salts at their present prices and the saving realized when sub- stituting sodium sulphate, so it is unnecessary to discuss this phase in this paper. Nine samples were used, namely: Dissolved hair waste; dissolved leather waste; dried blood; fish scraps; cottonseed meal; tankage; raw bone; 2 to 9 base goods; and calcium cyanamid. The following table shows the results obtained: ' Present address, Bureau of Chemistry, Washington, D, C. 2 J. Ind. Eng. Chem., 1916, 8: 586. 1920] JARRELL: AMMONIA IN FERTILIZERS 305 Comparison of sodium sulphate with potassium sulphate. AMMONIA FOUND USING She Potassium Sodium sulphate sulphate per cent per cent Wolo tHair'waste: se ee LP ae eee yee 11.44 11.44 11.34 11.36 11.36 11.40 (Average! 421 Bess 0 fy eat 2) |. 5. a ee aes 11.38 11.40 No. 2. Leather waste_____- MSS OIELS PE ew SEES Rae oh 9.48 9.48 9.46 9.45 9.58 9.52 Ay erare mene: crete. Choy Ser ie! se a 9.51 9.49 Nos ois pried) blood -t4 4. Vee 2.4 ).552> oe 2s tee epee 15.76 15.70 15.66 15.76 15.76 15.62 Avera perm lige IE eee) ge et)! 4) RS NE 2k 15.73 15.69 Ne weia? Wish scraps: tues emer sfla lia: FE eS 8.88 8.80 8.80 8.84 1] SSS GoRT aes. pepe a eat ee Se Sa ee ae 8.84 8.82 MasowGottonseedsumeal tg =e = 6.62 6.62 6.58 6.66 SRT aS PE a Ee ee ae 9 ee 6.60 6.64 oxo Iankeape! t! wt oe: ey (i) oak SE atte 9.08 9.18 9.08 9.07 Wveragel Seem. Teteeeys ite stetseb ese | FEL eB iets ty 9.08 9.13 INinmeemiit aw, DGHESesee se ee eee | a ee ee 4.47 4.54 4.45 4.35 4.52 4.47 Py erageseesee: &: Piney ise es beg bee bes 3 eae ge SE PPE Sos 4.48 4.45 Monte 2-9 Base poods = e+) 3.. Lae ane nee 2.25 2.27 2.22 2.24 PAwerngemee san sere fe fr 2th Ue eee eee 2.24 2.25 ea salem cyanamid.*=— =.= eee ee 15.22 15.26 15.22 15.28 15.32 15.28 ISG LT Se OE 29 ee ee = = ee ee 15.25 15.27 Meonerdlaveragets teh. 8 eo 1}. Sone eee ee 9.23 9.24 All of the determinations were made under the same conditions. The time of digestion was 2 hours and the digestion mixture consisted of 306 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 20 ce. of concentrated sulphuric acid, 10 grams of potassium sulphate or 10 grams of anhydrous sodium sulphate, and 1 gram of metallic mercury. The results show in every sample that sodium sulphate gives results practically identical with those obtained with potassium sulphate. The greatest difference in any one sample is well within the experimental error, being only 0.05 per cent of ammonia in Sample 6. The general average of all samples is 9.23 per cent when potassium sulphate is used and 9.24 per cent when sodium sulphate is used. In view of the above results, the writer concludes that sodium sulphate may be used as a substitute for potassium sulphate in the Kjeldahl- Gunning-Arnold method for the determination of ammonia in fertilizers. INVESTIGATIONS OF THE KJELDAHL METHOD FOR DETERMINING NITROGEN. By I. K. Puetps (Bureau of Chemistry, Washington, D. C.), Associate Referee on Special Study of the Kjeldahl Method, and H. W. Daupt* (Bureau of Chemistry, Washington, D. C.). The hydrolysis of pyridin compounds and other refractory substances was discussed in the report presented to this association in 1915. It was found that the hydrolysis of pyridin zinc chlorid, approximately 0.3 gram for each analysis, was complete when the digestion continued for 2} hours in an open Kjeldahl flask of 500 cc. capacity with a boiling mixture of 25 cc. of sulphuric acid, 0.7 gram of mercuric oxid and 10 grams of potassium sulphate. When sodium sulphate was substituted for an equal weight of potassium sulphate the results obtained were below the theory, the error equaling as much as 10 per cent of the total nitrogen. Return condensers, constructed entirely of lead, placed in the neck of the flask, reduced the amount of sulphuric acid volatilized during hydrolysis, causing a more constant proportion of sulphuric acid to potassium sulphate or sodium sulphate in the presence of mercury. These served not only to prevent the vaporization of sulphuric acid, but also to retain the acid ammonium sulphate even when excessive quanti- ties of potassium sulphate were employed. It was found that with return condensers constructed of lead the relative amounts of potassium sulphate and sulphuric acid in the presence of mercury determine the completeness of the hydrolysis. For instance, in the presence of 0.7 gram of mercuric oxid, 25 ec. of acid and 10 grams of potassium sulphate caused incomplete decomposition, but when 15 ce. of sulphuric acid were used with amounts of potassium sulphate varying from 15 to 30 ' Present address, Jackson Laboratory, E. I. DuPont Co., Wilmington, Del. 1920] PHELPS: INVESTIGATIONS OF KJELDAHL METHOD 307 grams, excellent results were obtained. When more potassium sulphate was employed the results were somewhat lower. Sodium sulphate seemed to give varying results. Recent investigations show that the use of disodium phosphate or of sodium pyrophosphate, with or without condensers, in the place of potassium sulphate is undesirable. Violent bumping usually occurred before the hydrolysis had progressed sufficiently. Further, the results were variable. Experiments conducted in open flasks with the boiling mixture of 10 grams of potassium sulphate, 25 cc. of sulphuric acid and various metallic catalysts showed that the hydrolysis of pyridin zinc chlorid is complete only when 0.7 gram of mercuric oxid is present. Mercuric oxid present in the proportion of 0.2 gram is insufficient for complete hydrolysis of so refractory a compound as pyridin, even when ideal proportions of sulphuric acid and potassium sulphate are used with condensers in the neck of the flask. Copper sulphate, nickel sulphate, potassium aluminium sulphate, zinc chlorid, manganese chlorid, manganese dioxid, tungstic acid, molybdic acid, titanic acid or vanadic acid under the conditions caused incomplete hydrolysis. It is to be noted that, in the presence of 0.7 gram of mercuric oxid and the proper proportion of potassium sulphate and sulphuric acid, hydrolysis is com- plete without the presence of copper sulphate. QUALITATIVE STUDY OF THE HYDROLYSIS OF AMINS. Methylamin, trimethylamin, cholin, betain and _ tetramethylam- monium salts were made and hydrolyzed at boiling temperature with suitable mixtures of sulphuric acid, copper sulphate, mercury, potas- sium or sodium sulphate. The products of hydrolysis were dissolved in water and potassium sulphid solution added if metallic catalysts had been used. This solution was made strongly alkaline with a saturated solution of sodium hydroxid and distilled into a slight excess of hydro- chloric acid. In a few instances the excess of N/5 hydrochloric acid was titrated with N/10 sodium hydroxid in order to determine the amount of alkali in the distillate. Methyl red was used as indicator. The determinations were made with H. E. Woodward of the Bureau of Chem- istry, according to the methods of Alsberg and Woodward! for detecting the presence of amins and trimethylamin. For this purpose the slightly acid solution was evaporated to a small volume and transferred to a test tube for the trimethylamin test. A normal solution of mercuric iodid in potassium iodid was carefully added from a graduated pipette until precipitation ceased. Each cubic centimeter of the reagent pre- cipitated approximately 0.06 gram of trimethylamin, the equivalent of 0.014 gram of nitrogen. When no trimethylamin was present the solu- tion was transferred to a small flask and distilled after making alkaline ‘Presented under the title “A New Reagent for Volatile Tertiary Amins” at the meeting of the American Chemical Society in New York, September 1916. 308 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 with sodium hydroxid. Sodium sulphid was added with the sodium hydroxid, if the solution had been tested with the trimethylamin reagent. From 5 to 10 cc. were distilled into 10 ce. of 40 per cent formal- dehyde solution in a test tube. Approximately 5 cc. of a N/4 solu- tion of mercuric bromid in potassium bromid was then added and the test tube warmed in the steam bath. The amount of amin present was estimated by the amount of precipitate. The hydrolysis of monomethylamin with sulphuric acid and potas- sium sulphate was found to proceed slowly unless the weight of potassium sulphate was equal to or greater than that of sulphuric acid. With sul- phuric acid and copper sulphate or mercuric oxid, the amin was even less completely hydrolyzed. When either copper sulphate or mercuric oxid was present with 25 ce. of sulphuric acid and 10 grams of potassium sulphate, the hydrolysis was rapid. In the presence of 1 gram of copper sulphate or 0.7 gram of mercuric oxid an amount of methylamin, con- taining 0.07 gram of nitrogen, was almost completely converted to ammonia by boiling for 1 hour. When 10 grams of sodium sulphate were used with either of the metallic catalysts, the hydrolysis was in general almost as effective as when potassium sulphate was similarly used. The hydrolysis of trimethylamin with sulphuric acid and potassium sulphate was found to proceed very slowly, even when the weight of potassium sulphate equalled that of the sulphuric acid. The action of either of the two metallic catalysts with 25 cc. of sulphuric acid and 10 grams of potassium sulphate was extremely slow. Trimethylamin was hydrolyzed very quickly by boiling with sulphuric acid, potassium sul- phate and either of the two metallic catalysts. The larger amounts of the catalysts were more effective than the smaller amounts. The effect of mercuric oxid in the presence of potassium sulphate and sulphuric acid seemed to be slightly impaired by the presence of chlorid, only when the hydrolysis was conducted for 1 hour or less with the 0.2 gram of the catalyst. A sublimate of mercuric chlorid was noted in a number of instances. When an equal weight of sodium sulphate was used in place of potassium sulphate, the hydrolysis was not so rapid, except when 0.7 gram of mercuric oxid was present. Complete conversion to ammonia was most quickly effected by the action at boiling temperature of a mix- ture of 25 cc. of sulphuric acid, 10 grams of potassium or sodium sulphate, and 0.7 gram of mercuric oxid, trimethylamin, represented by 0.05 gram of nitrogen, requiring less than 13 hours. Experiments were next conducted for the purpose of studying the hydrolysis of tetramethylammonium compounds, which from their con- stitution might be expected to yield trimethylamin. The amounts of tetramethylammonium chlorid hydrolyzed by boiling mixtures of sulphuric acid and copper sulphate or mercuric oxid were a ha @ tia vata 1920) PHELPS: INVESTIGATIONS OF KJELDAHL METHOD 309 very small. A mixture of 10 grams of potassium sulphate and 25 ce. of sulphuric acid was somewhat more effective, but the nitrogen recovered was only 70 per cent of that recovered when 0.7 gram of mercuric oxid and 10 grams of potassium sulphate were used. No positive reactions for trimethylamin were obtained in any of the experiments, while slight positive tests for amins were obtained when potassium sulphate and sulphuric acid were used, with or without mercuric oxid. Distillation with sodium hydroxid solution did not decompose an appreciable quantity of cholin. Hydrolysis with boiling sulphuric acid and copper sulphate or mercuric oxid took place very rapidly with the formation of large amounts of trimethylamin. Distillation from a strongly alkaline solution did not hydrolyze an appreciable amount of betain. The action of 10 grams of potassium sulphate with 25 cc. of sulphuric acid seemed to be effective whether the metallic catalysts were present or not. The action of sulphuric acid with copper sulphate or mercuric oxid was ineffective. No trimethyl- amin was found in any of the experiments with betain. Moderate amounts of primary or secondary amins were found, however. HYDROLYSIS OF CERTAIN ORGANIC COMPOUNDS. The hydrolysis of certain organic compounds of various constitutions was reported at the annual meeting of this association in 1915. Other substances have since been studied. In the presence of 0.7 gram of mercuric oxid, 10 grams of potassium sulphate and 25 cc. of sulphuric acid weights of the compound varying from 0.2 to 0.4 gram were hydro- lyzed completely by heating, without condensers, at the boiling point for 2% hours. The hydrolysis was found to be complete for the com- pounds grouped below. Glucosamin hydrochlorid. Isoquinolin derivatives: Tetramethylammonium derivatives: Papaverin. Tetramethylammonium iodid. Narcotin. Cholin hydrochlorid. Morphin. Pyrol derivative: Hydrastinin. Tsatin. Purin derivative: Pyrolidin derivatives: | Caffein. Atropin. Imidazole or glyoxalin derivatives: Cocain. Lophin. Pyridin derivatives: Amarin. Nicotin zinc chlorid. Histidin dibydrochlorid. Nicotinic acid. Quinoxalin derivative: Piperidin derivative: Quinoxalin hydrochlorid. 6-Eucaine hydrochlorid. Quinazolon derivatives: Quinolin derivatives: 2-Methy] 4-quinazolon. Hydroxyquinolin. 2-Methy] 3-phenyl 4-quinazolon. Cinchonidin. Strychnin. Brucin. 310 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 The above procedure, with such modifications as are necessary, was applied to certain compounds containing two nitrogen atoms directly united. THE DETERMINATION OF NITROGEN IN AZO COMPOUNDS. It was found that azo compounds were not completely hydrolyzed to ammonia by digestion with 0.7 gram of mercuric oxid, 10 grams of potas- sium sulphate and 25 to 30 cc. of sulphuric acid, whether applied directly or after preliminary treatment with zinc dust, salicylic acid and sulphuric acid, with zinc dust and sulphurous acid solution or with a mixture of fuming sulphuric acid and sulphur. When the preliminary treatment included solution in 20 cc. of alcohol and reduction with zinc dust and hydrochloric acid, the results were in fair accord with the theory. The hydrochloric acid reacted so slowly with the zinc that from 0.2 to 0.4 ce. of stannous chlorid solution, consisting of 40 grams of stannous chlorid in 100 ce. of concentrated hydrochloric acid, was added to hasten the action. The mixture was kept boiling for 15 minutes or for 7 minutes after decolorization. Glass return condensers of a modified Hopkins type were placed in the neck of the flasks to prevent the evaporation of alcohol. After cooling, an equal volume of water and 30 cc. of sulphuric acid were added and the mixture heated until the water had been expelled and foaming had ceased. After the addition of 0.7 gram of mercuric oxid and 10 grams of potassium sulphate the hydrolysis was conducted at the boiling point. Reduction with stannous chlorid in alcoholic solution was more efficacious. The azo compound was dissolved in 20 ce. of alcohol, 5 cc. of stannous chlorid solution, containing 40 grams of stannous chlorid in 100 cc. of hydrochloric acid, added, and the mixture kept at the boiling point for 15 minutes if bleaching is completed in 7 minutes or less. And, if bleaching required boiling for 15 minutes or more, boiling was continued for an additional 7 minutes. Return con- densers were employed to avoid the evaporation of alcohol. After cooling, an equal volume of water and 30 cc. of sulphuric acid were added and the mixture carefully heated until the water had been expelled and foaming had ceased. Then, after the addition of 0.7 gram of mercuric oxid and potassium sulphate, the hydrolysis was conducted at the boiling temperature. In some instances, where too much acid had been volatil- ized, the separation of stannous sulphate within the flask caused bumping. This was partly eliminated by heating after adding 5 cc. of concentrated sulphuric acid and heating until solution was again obtained. ‘The results agreed closely with the theory. The time required for discharging the color of the azo compound is much shorter if stannous chlorid is used for reduction than if zinc and hydrochloric acid are employed. Further, the results are slightly nearer the theory. 1920) PHELPS: INVESTIGATIONS OF KJELDAHL METHOD 311 The following compounds were investigated: Azobenzene. Diethyl red. Hydroxyazobenzene. Dipropyl red. Amidoazobenzene. Benzene azo, 8-naphthylamin. Toluene azo, p-toluidin. Ponceau 4 R. Methyl red. Congo red. DETERMINATION OF NITROGEN IN HYDRAZIN COMPOUNDS. It was observed that hydrazin sulphate and semicarbazid hydro- chlorid were not completely hydrolyzed by digestion with 0.7 gram of mercuric oxid, and 10 grams of potassium sulphate and 25 to 30 cc. of sulphuric acid, whether applied directly or after preliminary treatment with stannous chlorid or with zinc and hydrochloric acid; or, by forma- tion of hydrazin derivatives of glucose in the presence of sodium acetate and water; or by reduction of these compounds obtained as above with stannous chlorid, sodium amalgam or sodium formate. When zinc dust and acetic acid were employed, the results indicated complete hydrolysis. The procedure followed below seemed to be applicable to hydrazin, phenylhydrazin and phenylmethylhydrazin compounds, but not to semi- carbazid or oxamazid. The nitrogen compounds were dissolved in water; glucose, glacial acetic acid and zinc dust added in the order mentioned; and the mixture kept at the boiling temperature for 1 hour under a return condenser of the Hopkins type. Upon cooling, 30 cc. of concentrated sulphuric acid were added and the mixture carefully heated until the water had been evolved and foaming had entirely ceased. Then, after the addition of 0.7 gram of mercuric oxid and 10 grams of potassium sulphate, the hydrolysis was conducted at the boiling temperature. The following method was found to be of more general application. Alcoholic solutions of the nitrogen compounds were treated with for- maldehyde solution, zinc dust and concentrated hydrochloric acid. The mixture was kept at the boiling temperature for 30 minutes or more under a return condenser. After the reduction had progressed for 15 minutes a small quantity of stannous chlorid solution was added for the purpose of hastening the action of the acid on the zinc. After cooling, an equal volume of water and 30 cc. of sulphuric acid were added. When the water had been expelled the hydrolysis was conducted in the usual manner with mercuric oxid and potassium sulphate. The method - will give results in accord with the theory, if reduction of the aldehyde nitrogen complex to the amino compound is complete. It appears that the variability in the results is due to the impurity in the zinc dust rather than to the method. It seems to be a matter of some difficulty to obtain a supply of uniformly pure zinc dust. Oxid in the free metal, hydrolyzable nitrogen containing compounds, or both, present in the 312 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 zine dust in varying quantities, made it impossible to find a sufficiently uniform sample. It was noted that a proper state of subdivision was also an important factor in the efficacy of the zinc when employed as the reducing agent. Following is a list of the compounds investigated: Hydrazin sulphate. Phenylbenzoylhydrazin. Phenylhydrazin hydrochlorid. Diphenylbenzoylhydrazin. Bromphenylhydrazin. Semicarbazid hydrochlorid. Methylphenylbhydrazin sulphate. Phenylsemicarbazid. Diphenylhydrazin hydrochlorid. Oxamazid. p-Nitrophenylhydrazin. | FIG. 1. AERATION APPARATUS. A NEW AERATION APPARATUS. The addition of saturated sodium hydroxid to a strong acid solution is controlled and the reacting mixture kept agitated by means of a divided air current, the main portion of which serves to agitate the mixture, the auxiliary portion to force over the alkaline liquid. Undue 1920) PHELPS: INVESTIGATIONS OF KJELDAHL METHOD 313 violence of the reaction and the danger of drawing the absorbing solu- tion into the reaction flask are obviated. By partly closing the stop- cock controlling the main air supply, a portion of the air was diverted into the sodium hydroxid reservoir, thereby forcing that solution slowly into the flask. Furthermore, it seemed desirable to have a convenient method for handling and measuring the strong alkaline solution. The apparatus is shown in Figure 1. When the current of air is shut off at (a), sodium hydroxid solution is forced from the supply bottle (B) into the reservoir (C), the pressure being relieved through (b): After the liquid has reached a level in (C) higher than the top of (C,), the stop-cock (a) is opened, when the level automatically falls to the top of (C,). Now, (b) is partly closed, with the result that the sodium hydroxid in the reservoir is gradually forced into the flask through the tube (D,), while at the same time a current of air is passing through the same tube. By opening or closing (b), the sodium hydroxid can be added at will, drop by drop, if desirable. When all of the sodium hydroxid has been added, (b) is opened full and the aeration proper is begun. Pure ammonium sulphate was dissolved by gentle heating in various mixtures containing sulphuric acid, potassium sulphate, mercuric oxid and magnesium phosphate contained in a Kjeldahl flask. After water had been added to the contents of the flask and the apparatus connected, saturated sodium hydroxid was forced over and the ammonia aerated by means of a current of air approximating 850 liters per hour, into standard hydrochloric acid solution. A second aeration was made into fresh standard acid solution, and finally ammonia-free water was added to the contents of the reaction flask and a distillation made. In some instances the distillations were repeated after the addition of more ammonia-free water. Complete aeration and absorption of ammonia from 0.2400 gram of ammonium sulphate was secured in 25 minutes when from 15 to 20 cc. of sulphuric acid, 75 cc. of water and 125 cc. of saturated sodium hydroxid solution were employed. When 10 cc. or less of sulphuric acid were present, it was necessary to heat the mixture of sulphuric acid and water almost to the boiling point before adding the sodium hydroxid solution. When more than 25 cc. of sulphuric acid were present, complete aeration was not secured in the time indicated. The presence of as much as 0.7 gram of magnesium phosphate did not affect the results. When pure monomethylamin sulphate prepared from acetamid was aerated in a similar manner, it was found impossible to aerate com- pletely the volatile alkali. The equivalent of 0.0005 gram of nitrogen remained in the reaction flask and could be recovered by distillation after addition of ammonia-free water. When a mixture consisting of am- 314 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 3 monium sulphate and methylamin sulphate, in the proportions of their equivalent weights, and containing 0.3464 gram of nitrogen, was aerated in like manner, the aeration of the volatile acid consuming constituents was complete. It is to be noted that when hydrolysis is conducted with a boiling mixture of 25 cc. of sulphuric acid, 10 grams of potassium sulphate and 0.7 gram of mercuric oxid, less than 25 cc. of acid are present and more than 10 ce. should be present at the end of the digestion. Further, hydrolysis of amins to ammonia has been shown to be complete in less than 13 hours. Below is a summary of the advantages and disadvantages of the aeration method with the use of the above described apparatus. The sodium hydroxid solution is added to the mixture after the appa- ratus has been fully and completely connected. Bumping, which is so common when boiling strong alkaline solutions, particularly in the presence of insoluble matter and salts which separate out on boiling, is eliminated; also the use of zinc, which is added to prevent bumping. Furthermore, the injury and breakage of flasks caused by boiling strong alkaline solutions is reduced. The danger of boiling or spraying over is entirely eliminated, for an efficient device for breaking up the spray is utilized. The sodium hydroxid solution should be purified before use by aera- tion at the temperature of the steam bath, no blank for reagents being necessary in such a case. It is extremely difficult to obtain sodium hydroxid which will not yield an appreciable amount of volatile alkali on distillation. The handling of the strong sodium hydroxid solution is reduced to a minimum. Further, the solution does not come into con- tact with any stop-cocks or with rubber. If hydrogen sulphid is evolved when the sodium hydroxid is added, during aeration it is completely expelled from the standard acid solution. In the distillation process whatever hydrogen sulphid escapes absorp- tion by the alkaline solution is likely to remain in the acid solution and later affect the titration with certain indicators. The aeration can be stopped and then continued at any time if desir- able. The time of aeration can be made more definite than that of distillation. The time of actual distillation possibly can be made shorter than that of aeration, but it is done so with risk of mechanical transfer of sodium hydroxid or incomplete distillation of ammonia or both. However, the aeration method can be made to save time, because it is not necessary to cool the mixture of sulphuric acid and water before beginning the operation. Furthermore, the aeration method requires little attention between the time of starting and that of finishing. 1920] JARRELL: REPORT ON POTASH 315 Compressed air is not always available, but with few modifications the above apparatus can be adapted for suction. A good water pump will provide suction strong enough for several units. The use of burners and condensers is entirely eliminated. REPORT ON POTASH. By T. D. Jarret! (College of Agriculture, College Park, Md.), ‘ Associate Referee. The association at its last meeting recommended the following work for the determination of potash: (1) That further cooperative work on the perchlorate method for the determination of potash be discontinued for the present, but the suc- ceeding referee be advised to continue the investigation of the method with a view to perfecting the working details. (2) That further work be done on the method for obtaining water- soluble potash to determine whether hydrochloric acid shall or shall not be used before the precipitation with ammonium hydroxid and am- monium oxalate is made. The following instructions were sent to collaborators: Determine potash on as many of your own samples as possible by the following methods: (A) Official Lindo-Gladding Method. Use the process for preparing the water extract as adopted officially in 1912. (B) Modified Official Method. This is the same as (A) with the exception that the addition of 2 cc. of concentrated hydrochloric acid to water extract and boiling is omitted. After washing 2.5 grams on filter paper, add directly to the hot solution ammonium hydroxid and ammonium oxalate and proceed as in the official Lindo-Gladding method. PART I.—RESULTS OF COLLABORATIVE WORK. The following table presents a comparison of results obtained upon analyzing solutions of potash prepared according to the official method with those obtained on the same samples when solution was effected by the modified official method. 1 Present address, Bureau of Chemistry, Washington, D. C. ?U.S. Bur. Chem. Bull. 152: 41; 162: 48. 316 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TABLE 1. Effect of hydrochloric acid. AVERAGE AVERAGE NUMBER POTASSIUM POTASSIUM COLLABORATOR OF OxID OxID SAMPLES Ouricraa MODIFIED REPORTED METHOD OFFICIAL METHOD L. E. Westman, Laboratory of the Inland Revenue Berges Jin Bi! Department, Ottawa, Canada--------.------.-- 5 2.97 2.95 S. E. Asbury, College Station, Texas_----_----_--- 12 1.17 1.15 H. C. Moore, Armour Fertilizer Works, Atlanta, Ga. 10 3i5Y/ 3.57 J. D. Sessums, Agricultural College, Miss.__-------- 5 1.67 1.61 E. G. Proulx, Agricultural Experiment Station, La Hayetite;, Indsie iets tae up cel ee 8 8.93 8.86 Average for 40’samples_ __----------_-------- 3.66 3.65 The table gives the average results obtained by each collaborator, and a large range of fertilizer materials is represented. E. E. Vanatta (University of Missouri, Columbia, Mo.), working on sixty-six samples of commercial fertilizers, found thirty-four higher by adding hydrochloric acid, thirty higher by omitting hydrochloric acid, and no variation with two samples. DISCUSSION. The average of all results reported shows 3.66 per cent potassium oxid by adding hydrochloric acid to potash solution and 3.65 per cent potas- sium oxid by omitting it. Of the forty samples reported, twenty-four yield a slightly higher result by adding hydrochloric acid; thirteen, a slightly higher result by omitting hydrochloric acid. The results are the same on three samples. The difference between the two methods on every sample appears to be within the limit of the usual error of manip- ulation. The work of the past three or four years has shown that the addition of hydrochloric acid to the potash solution does not give higher results. It is recommended— That the official method for the preparation of potash solution in mixed fertilizers! be revised to read as follows: : Place 2.5 grams of the sample upon a 12.5 cm. filter paper and wash with successive portions of boiling water into a 250 cc. graduated flask until the filtrate amounts to about 200 ce. Add to the hot solution a slight excess of ammonium hydroxid and sufficient ammonium oxalate to precipitate all the lime present, cool, dilute to 250 cc., mix, and pass through a dry filter. ! Assoc. Official Agr. Chemists, Methods, 1916, 12. 1920] JARRELL: REPORT ON POTASH 317 PART I1—THE PERCHLORATE METHOD. For the purpose of testing the accuracy of the perchlorate method against potassium chlorid and potassium sulphate of tested purity, as well as mixtures of these salts with substances usually present in fertiliz- ers, solutions of the composition indicated in Table 2 were prepared. Analysis was performed according to the following methods: Method I. (Solutions A to I.) Place 20 cc. of the solution in an evaporating dish, add 5 cc. of perchloric acid (sp. gr. 1.12), evaporate on steam or sand bath until heavy fumes are emitted, take up the residue with 5 cc. of water, add 5 cc. of perchloric acid, and again evaporate until all free hydrochloric acid is driven off and dense white fumes of perchloric acid appear. _ (If the solution goes to dryness and a hard mass remains, take up with a few drops of perchloric acid.) When the evaporation is made on a water bath, place dish on hot plate and heat carefully until all hydrochloric acid is expelled. After cooling add 20 cc. of 95% alcohol and stir well. Allow to stand for 30 minutes. Decant the alcohol through a Gooch crucible having a fairly thick pad (4 inch) and wash twice by decanta- tion with 95% alcohol containing 0.2% perchloric acid!. Transfer the precipitate to crucible with the 95% alcohol containing perchloric acid and wash until the filtrate amounts to 75 or 80 cc. Finally, to wash out all perchloric acid, wash twice with alcohol-ether (1 part 95% alcohol to 1 part ethyl ether) using 3-5 ce. each time. Dry for 30 minutes at 120-130°C. and weigh. Dissolve the potassium perchlorate from the Gooch crucible with about 200 cc. of hot water and dry to constant weight in air oven. Cool and weigh. Loss in weight is potassium perchlorate. Method IT. (Solutions J and K.) Place 20 cc. of the solution in a porcelain or silica dish (do not use platinum), add an excess of 3% barium hydroxid solution and without filtering evaporate to dryness on sand bath. Ignite the residue below redness for 10 minutes over a Bunsen burner. Extract the residue with 20 cc. of boiling water, breaking up the material as much as possible. Filter into an evaporating dish of about 175 cc. capacity, and wash with boiling water until the filtrate amounts to 125-150 cc. Add 5 cc. of perchloric acid, evaporate carefully on sand bath until it fumes strongly, take up with 5 cc. of water, add a second 5 ce. of perchloric acid, evaporate, cool, and proceed in accordance with Method I. Method IIT. (Solution K.) Transfer 20 cc. to a platinum dish and proceed according to the official Lindo- Gladding method? until after the addition of 1 cc. of sulphuric acid (1 to 1) and ignition. Dissolve the residue in about 25 cc. of hot water, add about 2 cc. of concentrated hydro- chloric acid and add in slight excess a 10% barium chlorid solution acidified with hydro- chloric acid. Add the barium chlorid solution at the rate of about 1 drop per second. Filter? into an evaporating dish and wash the precipitate and filter paper thoroughly with hot water. Add perchloric acid and proceed as outlined in Method I. 1 Made by adding 1 ce. of perchloric acid (3p. gr. 1.12) to 100 cc. of 95% alcohol. 2 Assoc. Official Agr. Chemists, Methods, 1916, 13. 3 After precipitating with barium chlorid, it is often necessary to allow the hot solution to stand a short time to insure a precipitate which filters well and gives a clear filtrate. 318 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TABLE 2. Effect of added salls upon the determination of potash as potassium perchlorate. SOLUTION pee | POTASSIUM | POTASSIUM bronick oxi oxID (Made up to 1000 cc.) =a TAKEN FOUND gram gram gram A 0.3703 | 0.1263 | 0.1259 0.3717 | 0.1263 | 0.1264 Potassium chlorid, 10 grams 0.3707 | 0.1263 | 0.1260 0.3706 | 0.1263 | 0.1260 0.3711 | 0.1263 | 0.1262 0.3718 | 0.1263 | 0.1264 NV CRAG OS ee ree a a ne re eee | eee toes ol en eaee ay Potassium chlorid, 10 grams; sodi- | 0.3721 | 0.1263 | 0.1265 um chlorid, 15 grams 0.3706 | 0.1263 | 0.1260 Averages S. tak rere k= dll beep eee eee a Dea dant C } : Potassium chlorid, 10 grams; mag- | 0.3724 | 0.1263 | 0.1266 nesium chlorid, 10 grams 0.38720 | 0.1263 | 0.1265 BV CLARE aie a eke ll) ei | | Potassium chlorid, 10 grams; cal- | 0.3724 | 0.1263 | 0.1266 cium carbonate (dissolved in | 0.3697 0.1263 | 0.1255 dilute hydrochloric acid), 10 | 0.3711 0.1263 0.1262 grams 0.3725 0.1263 0.1267 IANVCV ARES ot jee cece ee NS ee | ee ees oe Potassium chlorid, 10 grams; di- te ‘i i sodium phosphate, 10 grams pete anes ae Averagen! le aS wei ver 540. i een eee & Bt ees Potassium chlorid, 10 grams; bari- | 0.3695 | 0.1263 | 0.1256 um chlorid, 10 grams 0.3696 | 0.1263 | 0.1257 ‘Average oy. Ju cua ht Sie Alc Ao i ehh BES aia A 1920) JARRELL: REPORT ON POTASH 319 TaBLeE 2.—Concluded. SOLUTION ao eee POTASSIUM | POTASSIUM ERROR ERROR CHLORATE OXID OXID POTASSIUM | POTASSIUM (Made up to 1000 cc.) SD TAKEN FOUND OXID OXID gram gram gram mq. per cent G 0.3716 | 0.1263 | 0.1263 0.0 100.00 0.3722 | 0.1263 | 0.1265 +0.2 100.16 Potassium chlorid, 10 grams; sodi- | 0.3699 | 0.1263 | 0.1258 —0.5 99.61 um chlorid, 15 grams; magne- | 0.3716 | 0.1263 | 0.1263 0.0 100.00 sium chlorid, 10 grams 0.3730 | 0.1263 | 0.1268 +0.5 100.40 LNIGhg 2 eee ae Ce el |e eee | eee ey | rane ene 100.03 H Potassium chlorid, 10 grams; sodi- | 0.3730 | 0.1263 0.1268 +0.5 100.40 um chlorid, 15 grams; magne- | 0.3703 | 0.1263 | 0.1259 —0.4 99.68 sium chlorid, 10 grams; calcium | 0.3730 | 0.1263 0.1268 +0.5 100.40 carbonate (dissolved in dilute | 0.3731 0.1263 | 0.1269 +0.6 100.48 hydrochloric acid), 10 grams 0.3708 0.1263 0.1261 —(012 99.84 moverapes set tt) Liew) tus Salt Sus 1s ite ST 2s 2262 100.16 J 0.1555 | 0.0541 | 0.0529 == 17) 97.78 0.1556 | 0.0541 | 0.0529 he, 97.78 Potassium sulphate, 5 grams; con- | 0.1563 | 0.0541 | 0.0531 Kt) 98.16 centrated hydrochloric acid, 10 | 0.1566 | 0.0541 | 0.0532 = 0:9) 98.34 ec.; barium chlorid in slight | 0.1571 | 0.0541 | 0.0534 ONT 98.71 excess 0.1554 | 0.0541 | 0.0528 a 97.60 USED pee Eee UES eye oo OE) ee eee Ie | eg aes 98.06 Js Potassium chlorid, 2.5 grams; po- | 0.1701 0.0586 0.0578 —0.8. 98.63 tassium sulphate, 2.5 grams; | 0.1724 | 0.0586 | 0.0586 0.0 100.00 magnesium sulphate, 2.5 grams; | 0.1732 | 0.0586 | 0.0589 +0.3 100.51 sodium chlorid, 5 grams; acid | 0.1717 0.0586 0.0584 =():2 99.66 phosphate (water extracted), 9 | 0.1729 | 0.0586 | 0.0588 +0.2 100.31 grams 0.1720 | 0.0586 | 0.0585 =O: 99.91 ASCE (ied 42 ES Se ED Be) ee (in =, ele |) ee a ae 99.84 K> 0.1704 | 0.0586 | 0.0579 =050, 98.81 Potassium chlorid, 2.5 grams; po- | 0.1698 | 0.0586 | 0.0577 Sue) 98.47 tassium sulphate, 2.5 grams; | 0.1705 0.0586 0.0580 —0.6 98.98 magnesium sulphate, 2.5 grams; | 0.1704 | 0.0586 | 0.0579 =a 7! 98.81 sodium chlorid, 5 grams; acid | 0.1713 | 0.0586 | 0.0582 —0.4 99.32 phosphate (water extracted), 9 | 0.1719 | 0.0586 | 0.0584 Oo 99.66 grams; concentrated hydrochlo- ric acid, 8 cc. AN CLARE ft Seen pense Bit eid 99.01 0.1705 | 0.0586 | 0.0578 —0.8 98.63 0.1723 | 0.0586 | 0.0586 0.0 100.00 0.1707 | 0.0586 | 0.0580 —0.6 98.98 0.1729 | 0.0586 | 0.0588 +0.2 100.31 0.1722 | 0.0586 | 0.0585 —0.1 99.91 0.1708 | 0.0586 | 0.0581 —0.5 99.15 A NODS Say TS EA leak ML as ale | de (i Lath sae 99.50 *Ammonium hydroxid and ammonium oxalate were not added to this solution. » Ammonium hydroxid and ammonium oxalate were added to this solution as per oflicial method. 320 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 DISCUSSION. The method used last year for washing the potassium perchlorate in the Gooch crucible has been modified. Since it was found that the use as a wash of 95 per cent alcohol saturated with potassium perchlorate often gives high results, due to deposition from the alcohol of some potassium perchlorate upon both the crucible and the potassium perchlorate, the washing was made in all determinations considered in this report with 95 per cent alcohol containing 0.2 per cent of perchloric acid; 50 to 60 ce. of alcohol were used and the precipitate washed twice with 3 to 5 ce. portions of alcohol-ether (1 to 1) to remove the last traces of per- chloric acid. The associate referee verified the findings of Davis! that 50 ce. of 95 per cent alcohol dissolve 0.0065 to 0.0085 gram of potassium perchlorate. By washing out the last traces of perchloric acid with alcohol-ether the loss is reduced to 0.0015 to 0.0020 gram of potassium perchlorate for each 50 ce. of wash. Alcohol containing 0.2 per cent of perchloric acid dissolves about 0.0015 gram of potassium perchlorate per 50 cc. It would appear, therefore, that the best results are to be obtained by a proper balancing of errors. In order that uniformity of results may be secured, a definite set of conditions must be followed most carefully. When washing with alcohol containing 0.2 per cent of perchloric acid, the quantity of alcohol used must be definitely set, and perchloric acid sufficient to combine with all bases present must be added. In a recent article Davis? has shown by experimental data that when potassium sulphate is evaporated directly with perchloric acid accurate results are obtained so long as perchloric acid remains in large excess. The associate referee used the procedure suggested by Davis without success, all results being very high. Under Solution I, Table 2, are shown the results obtained on potassium sulphate after precipitation of the sulphate with barium chlorid. All results are low, which is probably due to the occlusion of some potash by the barium sulphate. The barium hydroxid process (Method IT) has the following advantages over the barium chlorid process (Method IIT): (1) All ammonium salts are volatilized as ammonia. (2) Since magnesium salts are precipitated as magnesium hydroxid, the addition of sufficient perchloric acid to combine with all the magnesium is not required. (3) Soluble silicates are precipitated as barium silicate by the barium hydroxid. (4) The barium hydroxid method is shorter than treatment with sulphuric acid, burning and precipitation of sulphate with barium chlorid. 1J. Agr. Sci., 1912, 5: 64. 2 J. Chem. Soc., 1915, 107: 1678. Sa 1920] KUZIRIAN: ESTIMATION OF POTASSIUM 321 CONCLUSIONS. The perchlorate method for the determination of potash in potash salts and mixed fertilizers is quite accurate after the analyst has become acquainted with the details of manipulation. If sufficient perchloric acid is added, phosphates and sodium, mag- nesium and calcium salts produce no error. Sulphate and ammonium ions produce an error and must be removed before adding perchloric acid. After extracting the potash from mixed fertilizers with hot water, the addition of ammonium hydroxid and ammonium oxalate is not necessary. RECOMMENDATION. It is recommended that the referee next year study further the barium hydroxid process of the perchlorate method on mixed fertilizers of known potash content. The following is an incomplete bibliography of the recent literature on the perchlorate method: J. Am. Chem. Soc., 1914, 36: 2085. J. Agr. Sci., 1912, 5:. 52. Mining Eng. World, 1912, 36: 605. Z. landw. Versuchsw., 1915, 18: 77. J. Chem. Soc., 1915, 107: 361. Landw. Vers.-Sta., 1912, 78: 179. Tbid., 1915, 87: 365. Analyst, 1916, 41: 165. J. Chem. Soc., 1915, 107: 1678. THE SEPARATION AND GRAVIMETRIC ESTIMATION OF POTASSIUM. By S. B. Kuzreran' (Agricultural Experiment Station, Ames, Ia.). Sérullas? in 1831 proposed that the insolubility of potassium per chlorate in concentrated alcoholic solution be employed for the estima- tion of potassium. His method, however, received scant attention for the reason that no convenient method existed prior to 1912 for the preparation of perchloric acid. Recently Willard’ has developed a pro- cedure by which pure perchloric acid is produced with comparative ease. Cooperative work by T. D. Jarrell* indicated that the perchlorate 1 Present address, Box 87, Jamaica, N. Y. 2 Ann. chim. phys., 1831, 46: 294. 3 J. Am. Chem. Soc., 1912, 34: 1480. 4 J. Assoc. Official Agr. Chemists, 1915, 1: 400. 322 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 method in its present form is very unsatisfactory. Hill! has shown that anilin perchlorate, prepared easily from anilin oil and perchloric acid, has a definite composition and contains no water of crystallization. A known weight of the crystals dissolved in a measured quantity of abso- lute alcohol will, according to Hill, precipitate potassium quantitatively as perchlorate. A negative error of 0.0004 gram potassium oxid, or 1.5 per cent, was explained by Hill on the ground of an incomplete conyer- sion of potassium chlorid to potassium perchlorate. The best results are obtained by this method when the following precautions are taken. No alcohol more dilute than 99.5 per cent should be used; for every 1.5 cc. of water used for dissolving the mixed chlorids, 50 cc. of absolute alcohol should be added. A definite weight of anilin perchlorate is dissolved in 50 cc. of absolute alcohol and the solution added to the dissolved chlorids drop by drop with constant shaking and the mixture allowed to stand one hour before filtration. Under these conditions the following results were obtained: Precipitation of potassium® with anilin perchlorate. CORRESPONDING WEIGHT CORRESPONDING WEIGHT WEIGHT OF CALCULATED WEIGHT OF FOUND ERROR IN POTASSIUM POTASSIUM, |_—_————___—————|_ POTASSIUM CHLORID c. 5 PERCHLORATE 2 - OXID TAKEN Potassium Potassium FOUND Potassium Potassium (Loss) oxid perchlorate chlorid oxid gram gram gram gram gram gram gram 0.2005 0.1266 0.3726 0.3670 0.1975 0.1247 0.0019 0.2005 0.1266 0.3726 0.3685 0.1983 0.1252 0.0014 0.2000 0.1263 0.3717 0.3675 0.1978 0.1249 0.0014 0.2000 0.1263 0.3717 0.3670 0.1975 0.1247 0.0016 0.2000 0.1263 0.3717 0.3677 0.1979 0.1250 0.0013 0.2000 0.1263 0.3717 0.3676 0.1978 0.1249 0.0014 0.2000 0.1263 0.3717 0.3690 0.1986 0.1255 0.0008 0.2000 0.1263 0.3717 0.3680 0.1980 0.1251 0.0012 0.1000 0.0632 0.1858 0.1844 0.0992 0.0627 0.0005 0.1000 0.0632 0.1858 0.1840 0.0990 0.0625 0.0007 0.1000 0.0632 0.1858 0.1845 0.0993 0.0627 0.0005 0.1000 0.0632 0.1858 0.1843 0.0992 0.0627 0.0005 « The potassium chlorid used was recrystallized from the commercial C. P. product. When it was estimated as chloroplatinate, it showed a purity of 99.9 per cent potassium chlorid. 1 Am. J. Science, 1915, 4th ser., 40: 75. Sy ET se om 2 1920 ROSE: AVAILABILITY OF POTASH IN WOOD ASHES 323 The substitution of anilin perchlorate for perchloric acid shortens the process considerably. It also affords the best means for direct estima- tion of sodium in the filtrate. It has been the writer’s experience that some potassium chlorid is occluded by the perchlorate precipitate. Three series of four experi- ments each were conducted to establish this fact. When the precipitant, dissolved in the proper amount of alcohol, was added all at once and filtration completed within 15 minutes, the results were considerably below the theory. When, however, the precipitant was added drop by drop with constant shaking and the mixture allowed to stand about 2 hours before filtration, the results were decidedly better. Sulphates must be removed with barium chlorid. Unless the filtrate is to be used for the determination of sodium, the excess of barium chlorid does not interfere, provided enough of the precipitant is added to combine with all the bases present. In the opinion of the writer, anilin perchlorate is the best reagent to replace the highly expensive platinic chlorid. It is easily prepared and the manipulation is simple. For this reason it is recommended that further collaborative work be performed and that the latest work on the use of anilin perchlorate be thoroughly tested by the collaborators. A paper on the “Availability of Potash in Wood Ashes’! was presented by Messrs. R. E. Stallings and 8. H. Wilson of the Georgia Department of Agriculture, Atlanta, Georgia. A STUDY OF THE AVAILABILITY OF POTASH IN GCOMMER- CIAL WOOD ASHES. By R. E. Rosr (State Chemist, Tallahassee, Fla.). An unusually large number of samples of wood ashes has been analzyed by the Florida State Laboratory during the last two years. An examina- tion of the results shows that seldom is more than 4 per cent of water- soluble potash found; frequently less than 0.5 per cent is found. Table 1 exhibits a number of analyses. Among dealers and commercial and official chemists there has been considerable discussion of the methods employed to determine the availability or solubility of the potash in ashes. It has been claimed by some that the official method of this association for ashes—the water-soluble method—is unfair to the dealer or manufacturer, and that the Dyer method, 1 per cent citric acid solution? or a modification thereof would show a larger percentage of available potash. 1 Am. Fertilizer, 1917, 46: 24. 2H. W. Wiley. Principles and Practice of Agricultural Analysis. 2nd ed., 1908, 2: 533. 324 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TABLE 1. Determinations on samples of wood ashes. DESCRIPTION OF SAMPLE POTASH SAND CARBONATE OF LIME ComMeErcIAL Harpwoop ASHES pertcenE peneait op: Canadian hardwood ashes___-____-_-.-_-_---_-_- 2.72 32.73 63.64 Ashess @9 22). 2 Sa Che = a ey eg 1.47 27.99 70.50 @ypress (ashes: \aa==- 56 - a ee re = oe 0.61 19.27 78.22 Hardwood lasties>: San. 1seetiqneyiy Sale eies: 2.93 3:75 93.32 (Ashes 2S Bee ae ee ee 0.43 78.15 21.42 Gypresstashes=- ent see eee ee ee 0.36 18.20 80.84 Ashes 3-2) 2 -SaciShIaet “iam rate eae eye 0.90 28.26 70.84 NCU c Cerone ens See BS ee ae A eed 0.48 83.98 15.54 Hard wood ‘ashes=2>se0e- = oe en net eee ee 3.22 71.55 25.23 Hard wood*ashes! 24 a_ Sst - 40) erties. Sear iae ie 1.61 23.00 75.89 ard wood!ashesh 62s. f5 4.2 6 ee Bae 1.27 22.20 76.53 Nees INO. aos See Oe oo oo See ee nS ene! 3.37 8.02 88.61 INghes 28: {io Ry, Vo ney eS! & 3.29 9.00 87.71 Hardwoodyashesst 2s 8 4, Se ae 2) 0.86 46.16 34.78 UY 2 a a apt a5 8 A Ma gs acy Se Be 4.37 39.77 55.86 (Aghes? 3 teeny wr Viper: 24-12 ren k eeeetetler 3.42 20.66 76.00 IN GH eS Spb 55 ae eg 0.26 5.11 94.63 YN «Yeh pepe, ent a sea ae ey alent ace a o> mea 1.81 22.11 76.08 Ashes*NG: 22RtEL Ts TOLLE Rates PeeR yee ly ono 0.96 74.15 25.00 Ashes 30 ac oe os ne eee 0.28 23.20 76.52 DK: {ee ee cee ee 1.73 32.86 64.33 ComMMERCIAL PALMETTO ASHES Saw. Palmettovyashes:- 22. -_- .. 4224 .- Se 0.49 96.39 3.12 Palmetto mootmshes 3222-2" - 2 eee 0.24 94.20 5.56 Palmettotashes== =.= 2° 2 = 22. * Nee ene eee 107. | 2:2 oe Palmettorashes:)?-) 2h 9 6 ey er hae pep en! 0.25 96.30 2.90 Palmettovashes==- 222 ="- —- oe ee eee 1.44 °°)° °_ Se Palmetto rootiashes. 22 32- >= ee Peers REM Pee 4.04 $1.00 15.00 Palmeftoiashbes= ee) oa ee a. ae 2.33 77.20 20.47 Palmettotashes-% = 25 to vis tp in ee Te ee |e 0.5%) |v 3a Palmetto: ashes= 2) sae Oe eer ee eee cee 6.73 |" 2 a Palmettorashies!: 2. eae. = ok a ed 2:35: ||, 22a Palmettotashes. “st 278 Leni eb Wee iee te 3:35) 1/0 2 Se Ipalmnettoiashes: S12 2 22: 5 sn Be ete O51; |. 3 ee eee AVeragelel) 2 MIS. Siti l “ee 0 Mewes 1.86 89.20 9.41 In order to evaluate these claims, the Florida State Laboratory made a comparison of the methods for the determination of potash in wood ashes. Seven samples were analyzed by the following methods: (1) Official Lindo-Gladding Method'.—This method was employed with these modi- fications. A solution of platinic chlorid was used of such dilution that 1 cc. completely precipitated 1 per cent of potash (K»O) on the basis of a 1 gram sample. The weight of the potassium platinic chlorid was obtained by the difference between the weight of the crucible with the precipitate and that obtained after washing thoroughly with boiling water, followed by alcohol. 1 Assoc. Official Agr. Chemists, Methods, 1916, 12. 1920} ROSE: AVAILABILITY OF POTASH IN WOOD ASHES 325 (2) Hydrochloric acid-soluble potash.—Weigh 10 grams of the sample into a 500 cc. volumetric flask, add 50 ce. of hydrochloric acid (1 to 1), boil for 30 minutes, dilute with hot water to about 300 cc., add an excess of ammonium hydroxid and ammonium oxalate, and proceed according to the official method. (3) Dyer Method, potash soluble in 1% citric acid solution—Weigh 10 grams of the sample into a 500 cc. graduated flask, add 100 cc. of 1% citric acid solution. Stopper securely and digest at room temperature for 7 days with frequent shaking. Dilute to about 300 cc. with water, heat to boiling, add an excess of ammonium hydroxid and ammonium oxalate, and proceed as in the official method. (4) Modified Dyer Method—Use 250 ce. of 1% citric acid solution for each gram of sample and proceed in accordance with the Dyer method. TABLE 2. Comparison of methods of potash determination. ‘ HYDRO- MODIFIED | INSOLUBLE | GapRnonic DESCRIPTION CHLORIC | ASSOCIATION DYER DYER MATTER aCD OF SAMPLE ACID METHOD METHOD METHOD SAND Sane SOLUTION (SiOz) per cent per cent per cent per cent per cent Wood ashes____ 2.92 1.47 1.45 2.50 27.99 Large Palmetto ashes_ 0.23 0.24 0.23 0.19 94.20 Small Palmetto ashes_ 1.98 1.70 1.62 USP. 0 0] |e eee Small Wood ashes-_-__ 5.35 4.37 4.23 4.62 38.77 Large Palmetto ashes_ 4.71 4.73 4.60 3.14 81.00 Moderate Wood ashes____ 0.31 0.28 0.27 0.31 23.20 Large Wood ashes-___- 3.61 3.30 3.22 2.77 28.25 Large Feldspar® _____- 0.02 we 0.07 ONOG) fine eee yh Sess Average___ 2.73 2.30 2.23 2.12 49!07-9"i|) seeke. ee - * Not included in average. A study of the table indicates that the potash contained in commercial wood ashes is largely in the form of potassium carbonate and that very little, if any, exists in the form of silicate. This is shown in the samples having a large proportion of insoluble matter, largely silica, and a low percentage of potash. In these samples the tendency to form silicates would be greatest, and the water-soluble potash would be less than the acid-soluble. This, however, is not the case, the water-soluble potash agreeing with the acid-soluble in every case. CONCLUSIONS. A close comparison of the results obtained by using the four methods of solution shows: (1) Hydrochloric acid-soluble potash is uniformly higher than water- soluble. (2) Results for water-soluble potash are consistently higher than those obtained by the Dyer method; also results by the method of this association are much more concordant than those by the Dyer method. f 326 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 3 It appears that the potash in commercial fertilizers is not more soluble in dilute citric acid solution than in boiling water. Apparently nothing would be gained by substituting either the Dyer method or the modified Dyer method for the official procedure in the determination of the available potash in commercial wood ashes. The analytical work in connection with this paper was performed by Mr. L. Heimberger, fertilizer chemist of the Florida State Laboratory. No report was presented on soils by the referee. Messrs. T. E. Keitt and C. J. King (Agricultural Experiment Station of Clemson Agricultural College, Clemson College, 5S. C.) presented a paper! on “A New, Rapid and Accurate Method for Estimating Lime and Potash in Soils”. NITROGENOUS COMPOUNDS IN SOILS. By C. B. Lreman (Agricultural Experiment Station, Berkeley, Cal.), Associate Referee. A collaborative study of the determination of nitrogen in soils was made. The official Kjeldahl method for nitrogen in soils’, the official method for nitrogen in fertilizers’, and the Hibbard method, as modified by the associate referee on nitrogenous compounds in soils, were in- vestigated. Three main samples were chosen for collaborative work: (1) Light sandy soil; (2) light clay loam; (3) heavy brown clay adobe. Of each sample three subdivisions were made in the following manner: (A) Sifted through 1 mm. sieve and dried at 100°C.; (B) leached free from nitrates and dried at 100°C.; (C) sifted through 1 mm. sieve, 0.0200 per cent of nitrogen in the form of sodium nitrate added, carefully mixed and dried at 100°C. DISCUSSION OF RESULTS. The figures given in Table 1 are very disappointing in some respects and very illuminating in others. An almost invariable lack of agreement between duplicate and triplicate determinations, and also among numer- ous repetitions, is fully brought out in the extremes reported and empha- sizes the fact that no method herein studied seems to be accurate enough from the absolute standpoint to permit of the exact determina- tion of small changes in the nitrogen content of soils. Attention need only be called to the great discrepancy between the nitrogen content of 1S. C. Agr. Expt. Sta. Bull. 188. 2 Assoc. Official Agr. Chemists, Methods, 1916, 21. 5 [bid., 8. 1920) LIPMAN: NITROGENOUS COMPOUNDS IN SOILS 327 1A and 1B as determined by Mr. McLean and Mr. Roberts and the discrepancy frequently found between the determinations in the case of either analyst. These facts are not to be accounted for by lack of accuracy on their part, but merely emphasize the inherent differences between duplicate samples and the difficulty of duplicating digestion and distillation. TaBLe 1. Comparative results of nitrogen determinations in soils. | KJELDAHL METHOD HIBBARD METHOD MODIFIED KJELDAHL NUMBER ! re | BA |e] Bebe AES | Ma oes] Rak BS | aca] a | per cent) per cent) per cent| per cent'| per cent) per = per cent) per cenl|| per cent| per cent) per cent 1A = |0.0174|0.0167|0.0169/ 0.0270) |0.0139| 0.0157) 0.0186/0.0260 |0.0190/0.0200' 0.0270 0.0185) 0.0137) 0.0172/0.0260)|0.0210 0.0140) 0.0193! 0.9260)|0.0184/0.0214| 0.0270 1B __|0.0200/0.0162/0.0181/0.0280)|0.0165|0.0169/0.0176) 0.0280)|0.0209)0.0179| 0.0280 0.0197|0.0152| 0.0179) 0.0290) /0.0200' 0.0157 0.0225 |0.0290) 0.0217 0.0234|0.0250 1G _—_|0.0347|0.0277 | 0.0388/0.0390)|0.0301 0.0287|0.0295|0.0390)|0.0390)0.0300) 0.0450 0.0332/0.0322/0.0350/0.0390)}|0.0369)0.0304/0.0300/0.0390|!0.0405| 0.0353) 0.0440 2A = |0.0864)0.0792) © |0.0850)}0.0864)0.0815)0.0939/0.0850!| 0.0830) 0.0844/0.0930 0.0879/0.0762| © |0.0860}}0.0901|0.0801/0.0947/0.0860)| 0.0849) 0.0858 0.0760 2B _—_|0.0890|0.0822} © |0.0900|/0.0883)0.0791/0.0847| 0.0880) |0.0922)0.0877|0.0890 0.0892/0.0767; © |0.0910)|0.0914'0.0843 | 0.0906) 0.0890) 0.0926) 0.0906/0.0830 2C —0.1014/0.0932) © |0.0940)/0.0987|0.0969)0.1021/0.0940)/0.1081/0.1013)0.0970 0.1033)0.0912) © — |0.0950)/0.1032) 0.0913) 0.1043} 0.0940) /0.1076/0.1016)0.1000 3A |0.1149/0.1062} © |0.0990)|0.1097) 0.1008) 0.1164/0.1020)|0.1061)0.1121/0.1120 0.1138)0.1042) © |0.1010)|0.1195) 0.1043) 0.1160)0.1030)/ 0.1076 0.1146)}0.1120 3B |0.1149/0.1107} © /0.1110}/0.1169]0.1022/0.1115|0.1080}/0.1161/0.1121|0.1080 0.1144!0.1052) © |0.1130)/0.1203)0.1008)0.1141)0.1080)/0.1173)0.1171)0.1090 3C = /0.1265)0.1212; © 0.1140) 0.1267)0.1158/0.1286/0.1140) 0.1327) 0.1271/0.1250 0.1296|0.1192) © |0.1160)/0.1314/0.1190)0.1349|0.1150!/0.1327|0.1298|0.1320 * Agricultural Experiment Station, New Brunswick, N. J. > Agricultural Experiment Station, Lexington, Ky. © Agricultural Experiment Station, Knoxville, Tenn. @ Agricultural Experiment Station, Berkeley, Cal. * Satisfactory digestions not obtained because of going dry. There seems to be no great difference in the results obtained by the _ Kjeldahl and Hibbard methods. The Kjeldahl method modified to include nitrates gives sometimes better and sometimes poorer results than either of the other methods, showing higher results on soils to which nitrate has been added, but failing to recover quantitatively the added 0.020 per cent of nitrate nitrogen. 328 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [ Vol. J]I, No. 3 REMARKS. The associate referee deplores the situation with reference to the lack of cooperation or somewhat half-hearted cooperation offered by our agri- cultural chemists. This is not said in disparagement of the efforts of those who acted as collaborators, but with reference to those who have let the associate referee go on hoping for a year or more that the work would be completed and finally failed to submit any results. This condition perhaps is not due to anything wrong with individuals, but rather with experiment station organizations, and the associate referee must, therefore, urge upon you the necessity of adopting some different system for studying analytical methods. Many of the good soil chemists of today employ methods in their work which they have adopted without official test, but which seem to them superior, and, in all probability are far superior, to those which are in our books as official. The reason for this is that the association has not kept abreast of the great progress of soil science, and that we are of no assistance in the improvement of methods for the use of soil chemists. It is therefore suggested that our system of work as in vogue today be changed for purposes of expedition and increased output on the testing of methods by getting experiment stations officially to carry out the testing of methods as a regular research project, so that results may be called for and expected. The results do not, in the opinion of the associate referee, justify recommendations for the adoption of any new method, but he feels that the present official and unofficial methods for the determination of nitrogen in soils are unsatisfactory, and it is suggested that the official Kjeldahl method modified to include nitrates be discarded and that both the official Kjeldahl and Hibbard methods be further studied. It is also suggested that nitrate nitrogen be determined only by the colori- metric and reduction methods. RECOMMENDATIONS. It is recommended (1) That the association take steps towards haying the experiment stations carry out officially the testing of methods as a regular research project. (2) That further and more complete investigation be made of official and unofficial methods for the determination of nitrogen in soils. 1920) PATTEN: INORGANIC PLANT CONSTITUENTS 329 REPORT ON INORGANIC PLANT CONSTITUENTS: By Anprew J. Parren (Agricultural Experiment Station, E. Lansing, Mich.), Referee. The work of the past year has been confined, for the most part, to a study of McCrudden’s? method, which provides for the precipitation of calcium as oxalate in a boiling solution containing a small amount of hydrochloric acid. It was thought that McCrudden’s method of pre- cipitating calcium would overcome the error introduced by the occlusion of manganese, but this proved not to be the case. However, by titrating the calcium oxalate precipitate with standard permanganate solution, this source of error appears to be eliminated. Colorimetric methods for the determination of manganese have also been studied, and potassium periodate has been selected as the most satisfactory oxidizing medium. The color developed by its use is of the same shade as the standard, which, in our experience, is not always the case with other oxidizing agents. We used a composite solution made from the following substances in the proportions stated: per cent Phosphoric acid. (P:0;)2 2222204) 45.0 Galcimioxd) (Ga) = eee 3.0 Magnesium oxid (MgO)_______--_-_____ 10.0 Mangano-manganic oxid (Mn;0,)__-_____ 0.2 ern oxdi(be,0;)5----- - ee eee 1.5 Aluminic oxid (Al,03;)__________________ 1.3 IPotassiiim) Oxia (KsQ) 223 See SEE 25.0 Soitum oxi s(INasO) 5-2 = See 14.0 ‘Totals. ae eee. ne ee 100.0 The methods studied are as follows: CALCIUM. Dilute an amount of solution representing 0.5 gram of ash to 200 cc., add a few drops of alizarin and make slightly ammoniacal. Now add very dilute hydrochloric acid until the solution is faintly acid, followed by 10 cc. of N/2 hydrochloric acid and 10 cc. of 2.5% oxalic acid. Boil the solution until the precipitate becomes granular and add, with constant stirring, 15 cc. of a saturated solution of ammonium oxalate. Cool and add, with constant stirring, 8 cc. of 20% sodium acetate solution and allow to stand 4-18 hours. Filter and wash with hot water until free from chlorids. Dis- solve the precipitate in hot, dilute sulphuric acid and titrate with N/10 potassium permanganate solution. (1 cc. N/10 KMnO,=0.0028 gram CaO.) If preferred the calcium oxalate precipitate may be ignited and weighed, correcting for occluded manganese. 1 Presented by P. F. Trowbridge. 2 J. Biol. Chem., 1910, 7: 83. 330 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 MAGNESIUM. To the filtrate and washings from the calcium determination add 25 cc. of strong nitric acid and evaporate to dryness. Take up with dilute hydrochloric acid and make to volume of about 100 cc. Add 5 ce. of sodium citrate (10% solution) and 10 ce. of sodium hydrogen phosphate solution or enough to precipitate all the magnesium. Add dilute ammonium hydroxid, with constant stirring, until the solution is faintly alkaline, then add about 25 ce. of strong ammonium hydroxid and leave in a cool place overnight. Filter and wash with 2.5% ammonium hydroxid solution. Dissolve the precipitate in dilute hydrochloric acid and reprecipitate as before. Allow to stand 2 or 3 hours, filter and wash with 2.5% ammonium hydroxid solution, ignite and weigh as magnesium pyrophosphate. TABLE 1. Calcium and magnesium. CALCIUM MAGNESIUM REMARKS OXID OXID per cent per cent i: Ona Volumetric SOL ie aes B04) ees Volumetric SOE 1) 0) ae 1 gram of manganese sulphate added to solution before precipitating calcium 3.04 10.14 Gravimetric 3.12 10.10 Mangano-manganic oxid present, 0.24 per cent SOE | ees. Gravimetric Se: | eee Mangano-manganic oxid present, 0.20 per cent Average 3.055 10.12 Theory 3.00 10.00 FERRIC AND ALUMINIC OXIDS. To the filtrate and washings from the magnesium determination, add 25 ce. of con- centrated nitric acid and evaporate to dryness. Take up with dilute hydrochloric acid and dilute to about 150 ec. Make ammoniacal and allow to stand on the steam bath until the iron and aluminium have been precipitated as phosphates. Filter and wash with hot water until free from chlorids. Dissolve the precipitate on the filter with hot nitric acid (1 to 5), wash the filter thoroughly and precipitate as before. Filter, wash the precipitate with hot water until free from chlorids, ignite and weigh as ferric and aluminic oxids. If it is desired to determine the ferric oxid separately, fuse the residue with about 4 grams of potassium bisulphate, cool, add 5 ce. of concentrated sulphuric acid and digest until all sulphate is dissolved to a clear solution. Reduce with zinc, cool, and titrate with N/50 potassium permanganate solution. MANGANESE. To 50 ce. of the solution, representing 0.5 gram of ash, add about 15 ce. of concentrated sulphuric acid and evaporate to expel hydrochloric acid. When the solution has reached a small volume add 5-10 ce. of nitric acid and continue the evaporation. It is not 1920) ROARK: REPORT ON INSECTICIDES 331 necessary nor desirable to evaporate until dense fumes appear, since the ferric sulphate then dissolves with difficulty. Nitric acid may be present, but not hydrochloric. Add water, a little at a time, heat until the iron salts have dissolved and dilute to about 150 ce. Add about 0.3 gram of potassium periodate, heat just to boiling for a few minutes and allow to cool. The standard is prepared in the following manner: To a volume of water equal to the sample add 15 cc. of sulphuric acid and sufficient pure ferric nitrate, free from manganese, so that this solution will contain about the same amount of iron asthe samples. Add standard permanganate solution, noting the amount, until the color is slightly darker than the sample, and then the same amount of periodate and boil as before. When cool, transfer the sample and standard to 250 cc. flasks and make to mark. If the color is weak it may be necessary to make to smaller volume. Compare the colors in any standard colorimeter. TABLE 2. Manganese. MANGANG~ SAMPLE MANGANIC OXID icone we as 0.20 RECOMMENDATIONS. It is recommended— (1) That further study be made of the methods as outlined for calcium, magnesium, iron and aluminium with solutions approximating the com- position of the ash from cereals. (2) That further study be made of the colorimetric method for the determination of manganese. REPORT ON INSECTICIDES. By R. C. Roark! (Bureau of Chemistry, Washington, D. C.), Referee. The work included a study of methods for the determination of arsenic trioxid and arsenic pentoxid in the presence of each other in lead arsenate; for the determination of lead oxid, zinc oxid and copper in such preparations as Bordeaux-lead arsenate, Bordeaux-zine arsenite, etc.; and for the analysis of lime-sulphur solution. * Present address, General Chemical Co., Baltimore Works, Baltimore, Md. 332 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 LEAD ARSENATE WITH LEAD ARSENITE. To test methods for the direct determination of trivalent and pen- tavalent arsenic in the presence of each other, a sample of lead arsenate with lead arsenite was prepared by thoroughly mixing 300 grams of dilead arsenate (PbHAsO,) with 100 grams of lead arsenite. The lead arsenate was precipitated by the addition of a solution of lead nitrate to a solution of potassium dihydrogen arsenate (KH.AsO,), the latter being in slight excess. The lead arsenite was prepared by adding a solution of arsenic trioxid in water to a solution of lead acetate contain- ing a little free acetic acid. The slight amount of precipitate that formed on standing overnight was filtered off, and ammonia added in excess to the filtrate, throwing down lead arsenite, Pb;(AsO;)2. Both the lead arsenate and lead arsenite were thoroughly washed, dried, and put through a No. 40 sieve before mixing. On analysis the lead arsenate was shown to have the theoretical com- position for dilead arsenate (33.11 per cent of arsenic pentoxid), while the lead arsenite contained 21.04 per cent of arsenic trioxid and 0.14 per cent of arsenic pentoxid. The mixture of the two submitted for cooperative work should contain, therefore, 5.26 per cent of arsenic trioxid and 24.86 per cent of arsenic pentoxid, or a total of 30.97 per cent calculated as arsenic pentoxid. The methods tested on this sample are as follows: TOTAL ARSENIC TRIOXID. Weigh an amount of the powdered sample equal to the amount of arsenic trioxid to which 1000 cc. of the iodin solution are equivalent. Transfer to a 200 cc. graduated flask, add 100 cc. of dilute sulphuric acid (water, 85 cc.; concentrated sulphuric acid, 15 ec.), and boil for 30 minutes. Cool, make to volume, shake thoroughly, filter through a dry filter, take 100 cc. of the filtrate and nearly neutralize with a strong solution of sodium hydroxid, using a few drops of phenolphthalein as indicator. If the neutral point is passed make acid again with dilute sulphuric acid, then add sodium bicarbonate in excess and titrate with N/20 iodin solution in the usual way. The number of cc. of iodin solution used in this titration multiplied by 0.2 equals the per cent of arsenic trioxid in the sample. TOTAL ARSENIC PENTOXID. REAGENTS. Solutions required: Starch solution—Prepare as directed under Paris green'. Standard iodin solution—Prepare as directed under Paris green, but calculate in terms of arsenic pentoxid. Standard thiosulphate solution.—Prepare an approximately N/20 solution as follows: Weigh 13 grams of crystallized C. P. sodium thiosulphate, dissolve in water which has been recently boiled and cooled, filter, and make to volume in a 1 liter graduated flask, using water that has been recently boiled and cooled. To standardize this solu- tion, proceed as follows: ' Assoc. Official Agr. Chemists, Methods, 1916, 63. 1920) ROARK: REPORT ON INSECTICIDES 333 (a) Dissolve about 0.7 gram of C. P. dilead arsenate (PbHAsO,!) in 50 cc. of con- centrated hydrochloric acid in an Erlenmeyer flask. If necessary to effect solution, heat on the steam bath, keeping the flask covered with a watch glass to prevent evapora- tion of the acid. Cool to 20-25°C., add 10 cc. of potassium iodid solution (20 grams of potassium iodid per 100 cc.), and 50 cc. (or more if necessary to produce a clear solution) of ammonium chlorid solution (25 grams of ammonium chlorid per 100 cc.) and immedi- ately titrate the liberated iodin with the thiosulphate solution. When the color of the solution becomes a faint yellow, dilute with about 150 cc. of water? and continue the titration carefully, drop by drop, until the solution is colorless, using starch paste as an indicator near the end point. From the weight of lead hydrogen arsenate and the number of cc. of sodium thiosulphate solution used, calculate the value of the latter in terms of arsenic pentoxid. . (AsO; in PbHAsO,=33.11 per cent.) (b) Titrate 50 cc. of the standard iodin solution, diluted with about 100 cc. of water, with the thiosulphate solution, to a colorless solution, using starch paste as an indicator, and from the ratio of the two solutions, and the value of the iodin solution in terms _of arsenic pentoxid, calculate the value of the thiosulphate solution in terms of arsenic pentoxid. The values obtained by these two methods of standardization should check very closely. The value obtained by procedure (a) is to be preferred. DETERMINATION. Weigh an amount of the powdered sample equal to twice the amount of arsenic pentoxid to which 100 cc. of the thiosulphate solution are equivalent. Transfer to an Erlenmeyer flask, dissolve in 50 cc. of concentrated hydrochloric acid, and proceed as directed under standardization (a). The number of cc. of thiosulphate solution used in the titration, divided by 2, represents directly the per cent of arsenic pentoxid in the sample. The following results on this sample have been received: 1 Pure dilead arsenate may be prepared by pouring a solution of lead nitrate into a solution of potas- sium dihydrogen arsenate (KH2AsO,), which should be in excess. The precipitate should be collected by filtration, dissolved in the smallest possible quantity of boiling nitric acid (1 to 4), and this solution pe penned ae = large quantity of distilled water. The precipitate which results should be collected an at 11 * Later results indicate that dilution is unnecessary. It is very important to standardize and analyze under the same conditions, especially concentration of acid. 334 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 Determination of arsenic triorid and arsenic pentorid in mixtures of lead arsenate and lead arsenile. TOTAL TOTAL ANALYST muoxip | rentoxrp | (ARSENIC |permitaTioN, (As20s3) (As2O5) As As:Os ae nt nt nt nt J. J.T. Graham, Bureau of Chemistry, | "4092 | 2533 | —-. | 3005 Washington, D. C. 4.87 25.30 pe Bed 2 31.00 4.85 25.20 ee a 4.82 ofae eats ee 4.90 [esses ree sane ‘Averages -42= = Stee. Sree oe 4.87 25.28 30.94 30.98 A. J. Flume, Agricultural Experiment 4.68° 25.69% eae bee, Station, Geneva, N. Y. 4.60* 25.748 See 30.10 4.75» 26.10” aS =e 4.67> 26.15> aie 30.53 4.64° 25.51° nee, E 4.58° 25.56° BPLe 29.89 Averages <4. 5 ate rs, ee 4.65 25.79 31.19 30.17 H. L. Fulmer, Ontario Agricultural 5.12 25.7 31.65 = = ee College, Guelph, Canada. D. K. French, Dearborn Chemical Co., 5.52 24.81 Bir 31.24 Chicago, Il. 5.52 24.85 Ysa 31.25 5.51 24.79 en ae 31.27 Averages tsi ee ber eee 5.52 24.82 31.24 31.25 R. C. Roark, Bureau of Chemistry, 5.32 25.10 =ee 31.26 Washington, D. C. 5.26 25.10 Pens = oe AVeri Ses: 2<- ons -seebenceeeeees 5.29 25.10 31.25 31.26 R. N. Miller, Bureau of Chemistry, 5.32 25.10 =som 31.28 Washington, D. CG. 5.24 2p she 31.13 5.28 bgey es 3 - Meese IAVCTARC = = os ore ee 5.28 25.13 31.26 31.21 W. W. Webber, Agricultural Experi- 4.82 25.76 Sa eee ment Station, Orono, Me. 4.88 25.60 ot 58 wise Average! con2 202 ee oss ter eS 4.85 25.68 31.31 Se 2 General average-_-_--------------- 5.00 25.40 31.21 30.90 Calculated value- ---.-.-.------- 5.26 24.86 30.97 30.97 * Standardized with arsenic trioxid. > Standardized with dilead arsenate. e Standardized with potassium iodate. COMMENTS BY ANALYST. A. J. Flume.—The thiosulphate solution was standardized by means of dilead arsenate and potassium iodate. A ratio was established with the iodin solution, and the arsenic equivalent of both solutions was determined. The iodin solution was standardized with arsenic trioxid, and from the ratio above the arsenic equivalent of the thiosulphate was calculated. 1920] ROARK: REPORT ON INSECTICIDES 335 DISCUSSION. There is quite a variation in the results for both arsenic trioxid and arsenic pentoxid as determined by the different analysts. This is no doubt due to incomplete mixture of the sample before weighing out portions for analysis. The lead arsenate was a very fine, light and fluffy powder, while the lead arsenite was dense and in the form of much coarser particles, which had a tendency to separate on agitation. The referee found it necessary to mix thoroughly each sample in order to get duplicate portions from the same bottle to agree. The fact that for each analyst the sum of the arsenic trioxid and arsenic pentoxid, calculated to a common basis, agrees with the total arsenic determined by the distillation method, shows that the methods for these determinations are accurate. - Thereferee presents in a separate paper, pp. 365—7, the results of analy- sis of a number of commercial lead arsenates by the above method for arsenic trioxid and a slight modification of the method for arsenic pentoxid, showing that these methods yield accurate results on commer- cial samples. The fact that one of the analysts obtained low results for total arsenic by the distillation method may be accounted for by the presence of a large amount of cupric chlorid in the cuprous chlorid used. Samples of so-called cuprous chlorid were found sometimes to consist almost entirely of cupric chlorid'. BORDEAUX-LEAD ARSENATE AND BORDEAUX-ZINC ARSENITE. For testing the methods for lead oxid, copper and zine oxid, two samples were prepared: No. 1, Bordeaux-lead arsenate; No. 2, Bordeaux zine arsenite. The Bordeaux-lead arsenate was prepared by mixing 280 grams of dilead arsenate, made as previously described but recrystallized from nitric acid, with 140 grams of dry Bordeaux mixture, made from C. P. materials. This Bordeaux, when analyzed by the official electrolytic method, showed a copper content of 18.45 per cent, and the lead arsenate yielded theoretical results for lead oxid in dilead arsenate, namely, 64.29 per cent. The Bordeaux-lead arsenate should, therefore, contain 21.43 per cent of lead oxid and 12.08 per cent of copper. 1A solution of cuprous chlorid may be prepared by dissolving cupric chlorid in hydrochloric acid (sp. gr. 1.19), adding several strips of copper foil and enough water to prevent loss of hydrochloric acid when the solution is boiled, and heating for an our. An excess of metallic copper must always be present. The solution is then poured into about twice its volume of water, the cuprous chlorid which precipitates is collected on a Biichner filter, washed quickly with alcohol, dried at 110°C., powdered and preserved in a glass-stoppered bottle in the dark. By removing the water from the cuprous chlorid with alcohol, no oxidation occurs on drying. Copper oxid or copper carbonate may be used instead of cupric chlorid for preparing cuprous chlorid in the above manner. he filtrate from the precipitated cuprous chlorid still contains a considerable amount of this salt. It may be concentrated by evaporation in the presence of metallic copper and more cuprous chlorid obtained as before. 336 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 In preparing the Bordeaux-zine arsenite, 200 grams of the above Bordeaux were mixed with 200 grams of zinc meta-arsenite, ZnAsoOx. This latter was prepared according to the method of Avery! by adding a solution of arsenic trioxid to a solution of zinc chlorid, both solutions being slightly acid, and then adding sodium hydroxid solution to neu- trality. A determination of zinc in this sample of zinc meta-arsenite by the method of Balls and McDonnell? yielded 23.45 per cent of zinc, equivalent to 29.19 per cent of zinc oxid (theory 29.13 per cent). The Bordeaux-zinc arsenite should contain, therefore, 14.57 per cent of zinc oxid and 9.24 per cent of copper. The directions that were sent out for the determination of these con- stituents are as follows: GENERAL PROCEDURE FOR THE ANALYSIS OF A PRODUCT CONTAINING ARSENIC, ANTIMONY, LEAD, COPPER, ZINC, IRON, CALCIUM, MAGNESIUM, ETC. (Applicable to such preparations as Bordeaux-lead arsenate; Bordeaux-zinc arsenite; Bordeaux-Paris green; Bordeaux-calcium arsenate, etc.). Lead oxid (PbO).—Weigh 1 gram of the dry powdered sample, transfer to a beaker, add 5 ec. of hydrobromic acid (sp. gr. 1.31) and 15 ce. of hydrochloric acid (sp. gr. 1.19) and evaporate to dryness to remove arsenic; repeat this treatment; then add 20 cc. of hydrochloric acid (sp. gr. 1.19) and again evaporate to dryness. Dissolve in 25 ce. of 2N hydrochloric acid, dilute to 100 cc. and pass in hydrogen sulphid until precipi- tation is complete. Filter, and wash precipitate thoroughly with N/2 hydrochloric acid, saturated with hydrogen sulphid. Save the filtrate and washings for the deter- mination of zinc. (If antimony is present it should be removed by digesting the sul- phids with sodium sulphid. However, antimony is only occasionally present and always in small amount and usually only in samples containing zinc, so that in general this step may be omitted.) Transfer the filter paper containing the sulphids of lead and copper to a porcelain casserole or evaporating dish and completely oxidize all organic matter by heating with a few cc. of concentrated sulphuric acid, together with a little fuming nitric acid; then completely remove all nitric acid by heating on the hot plate with sulphuric acid to copious evolution of white fumes, cool, and determine lead as the sulphate as directed for lead arsenate’. From the weight of lead sulphate calculate the amount of lead oxid present, using the factor: PbSO,X0.73600 =PbO. Copper.—Evaporate the filtrate and washings from the lead sulphate precipitate, remove excess of sulphuric acid by fuming on the hot plate, and determine copper as directed under Bordeaux mixture, either by the electrolytic or Low’s titration method*. Zine.—Evaporate the filtrate and washings from the precipitate of copper and lead sulphids to a small volume, add 1 or 2 cc. of concentrated nitric acid, boil for a few minutes, then evaporate to dryness, add a few cc. of concentrated sulphuric acid and heat to fuming, dissolve in a little water, neutralize with concentrated potassium hydroxid solution, using phenolphthalein solution as indicator, and then add not over 10 grams excess solid potassium hydroxid. ‘Transfer to a weighed nickel crucible and electrolyze, using a rotating anode and a current of about 3.5 amperes. When deposi- tion is complete, which should take about 1 hour, wash with water by siphoning, then 1J. Am. Chem. Soc., 1906, 28: 1163. 2J. Ind. Eng. Chem., 1915, 7: 26. 3 Assoc. Official Agr. Chemists, Methods, 1916, 68. * [bid., 70. 1920] ROARK: REPORT ON INSECTICIDES 337 rinse with alcohol, dry for a minute or so in an oven, and weigh as metallic zinc. Cal- -culate zine oxid, using the factor: Zn 1.24476 =ZnO. Results on the samples of Bordeaux-lead arsenate and Bordeaux-zinc arsenite received are as follows: Cooperative results. ANALYST O. D. Knight, Bureau of Chemistry, Washington, D. C. Ate Sa a Se D. K. French bp STERTOR as 5 te cy re eg elt R.N. Miller Le SEITE gee dm le eee aah Sing R. C. Roark ' _N. H. Borden, Bureau of Chemistry, Washington, D. C General Average________________ Calculated value__.___.________- BORDEAUX-LEAD ARSENATE Lead oxid (PbO) per cent 21.27 21.27 21.27 21.37 21.27 21.37 21.31 21.33 19.74 19.52 19.52 19.59 21.26 21.20 21.19 i) apa ' orb ' wh bo ra wo “I a Se) Se No bot wWm~I19 BORDEAUX-ZINC ARSENITE “ey || ao | aS per cent per cent per cent 12.08 13.49 9.33 12.10 13.53 9.33 12.09 13.51 9.33 12.32 12.12 9.26 12.26 12.32 9.34 12.30 sae 9.23 12.14 Sree 9.28 12.26 12.22 9.28 11.42 _ 2 oe 11.48 = ssek 11.45 aoe a ee 13.18 pes 9.78 13.20 Hye 9.75 13.03 =e 9.86 13.14 _— 9.80 12.28 13.17 9.34 12.24 14.09 9.38 es 11.30 _¥s 12.26 12.85 9.36 11.71 13.36 9.14 11.60 13.82 9.24 11.64 nr ae re 11.65 13.59 9.19 12.30 ——— 9.912 12.30 eye! 9.865 12.30 S322 9.89 12.37 13.98 9.28 12.39 13.98 9.28 12.38 13.98 9.28 12.22 13.20 9.45 12.08 14.57 9.24 * Low's titration method. » Electrolytic method. 338 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 COMMENTS BY ANALYSTS. J. J.T. Graham.—In the method for lead oxid in Bordeaux-lead arsenate it is suggested that the referee’s instructions be amended to eliminate any silica that may be present, as this may cause an error. N. H. Borden.—In the electrolytic determination of zinc 14.17 per cent of zine oxid was obtained, using 10 grams excess of sodium hydroxid instead of potassium hydroxid. In all cases the electrolysis was continued about 75 minutes. DISCUSSION. The method of getting rid of arsenic by reduction to As™ with hydro- bromic acid and volatilization of the trichlorid by evaporation of the hydrochloric acid solution to complete dryness works very satisfactorily. The precipitation of the sulphids of lead and copper is accompanied with certain difficulties. If the solution is too acid the precipitation of lead is not complete; if not sufficiently acid, some zinc sulphid is brought down. The results for lead oxid obtained by the analysts are good, being low in one instance; the results for copper are only fair; and the results for zinc oxid are quite poor. It is recommended that more work be done on these or other methods for the determination of these substances before adoption by the association. LIME-SULPHUR SOLUTION. To test the method for lime-sulphur solution, three samples were sent out. Sample 1 was prepared by passing hydrogen sulphid into a mixture of C. P. calcium hydroxid and water for several days, boiling this solu- tion of calcium hydrosulphid (together with a slight excess of calcium hydroxid remaining unacted upon) with an excess of sulphur in a large flask provided with a reflux condenser, a stream of hydrogen sulphid being passed during the operation. When the reaction had proceeded to apparent completion the solution was filtered hot, allowed to cool in the absence of air and bottled. Sample 2 was prepared by mixing 1100.5 grams of Sample 1 with 999.0 grams of a solution of sodium thiosulphate prepared by dissolving 400 grams of C. A. F. Kahlbaum’s sodium thiosulphate in water and making the solution up to 1000 ce. Analysis of this solution of sodium thiosulphate showed thiosulphate sulphur by titration with iodin solu- tion equivalent to 8.61 per cent. Sample 2 should contain, therefore, 4.097 per cent of thiosulphate sulphur plus 0.52 times the thiosulphate sulphur in Sample 1. It is extremely unlikely, considering the method of preparation, that Sample 1 contains as much as 0.10 per cent of thiosulphate sulphur, and certainly not more than this amount. Assum- ing 0.10 per cent of thiosulphate sulphur for Sample 1, Sample 2 should contain 4.15 per cent of thiosulphate sulphur. Sample 3 is a commercial concentrated lime-sulphur solution. 1920] ROARK: REPORT ON INSECTICIDES 339 The following directions were sent out: PREPARATION OF SAMPLE. Proceed as directed in the Association of Official Agricullural Chemists, Methods, 1916, 76. TOTAL SULPHUR. Proceed as directed in the Association of Official Agricullural Chemists, Methods, 1916, 76. I. Zine Chlorid Method. SULPHID SULPHUR. Pipette 10 cc. of the solution prepared for analysis into a beaker, and immediately add a slight excess of ammoniacal zinc solution (prepared by dissolving 50 grams of pure zinc chlorid in water, adding ammonia in sufficient quantity to redissolve the precipitate first formed, and then adding 50 grams of ammonium chlorid! and diluting to lliter). Stir thoroughly so as to assure complete precipitation of the sulphid sulphur, filter, and wash 4 or 5 times with hot water. Transfer the filter paper containing the zinc sulphid to the beaker in which the precipitation was made, cover with a little water, disintegrate with a glass rod, and add about 3 grams of sodium peroxid, keeping the beaker well covered with a watch glass. Warm on the steam bath with frequent shaking until all the sulphid is oxidized to sulphate, adding more sodium peroxid if necessary. Make slightly acid with hydrochloric acid, filter to remove shreds of filter paper, wash thoroughly with hot water, and determine sulphur in the filtrate exactly as directed under “total sulphur’. THIOSULPHATE SULPHUR. Proceed as directed in the Association of Official Agricultural Chemists, Methods, 1916, 76. Use very little starch paste in the titration so as not to interfere with the precipitation of sulphate sulphur with barium chlorid. SULPHATE SULPHUR. Proceed as directed in the Association of Official Agricultural Chemists, Methods, 1916, 77. II. Cadmium Chlorid Method. Proceed as directed in the zinc chlorid method, substituting for the ammoniacal solu- tion of zinc chlorid an ammoniacal solution of cadmium chlorid prepared as follows:* Dissolve 50 grams of cadmium chlorid in water, add 400 cc. of ammonia (sp. gr. 0.90) and dilute to 1000 ce. IIT. Todin Titration Method. For preparation of sample for analysis, determination of monosulphur equivalent and thiosulphate sulphur by methods “A’’, ““B”, “C’’, and “D’’3, use N/20 iodin solu- tion standardized against pure arsenic trioxid and calculate results as follows: As»O; X 1.29588 =thiosulphate sulphur (S). As,O; X 0.32397 =monosulphur equivalent sulphur (S). TOTAL LIME. Proceed as directed in the Association of Official Agricultural Ciemists, Methods, 1916, 77. Observe the precautions about solubility of calcium oxalate in hot water and directions for ignition given by Hillebrand.‘ Results received are as follows: 1J. Soc. Chem. Ind., 1912, 31: 369. 2 Thompson and Whittier. Del. Agr. Expt. Sta. Bull. 105: 7. 8 J. Ind. Eng. Chem., 1916, 8: 624. 4U.S. Geol. Surv. Bull. 422: 119. 340 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 ANALYST J. J. T. Graham Average F. C. Cook, Bureau of Chemistry, Washing- ton, D. C Average S. D. Averitt, Agricultur- al Experiment Station, Lexington, Ky. Average D. K. French Average R. M. Chapin, Bureau of Animal Industry, Washington, D. C Average C. H. Robinson, Experi- mental Farm, Ottawa, Canada. AV Orage: = see aes O. B. Winter, Agricul- tural Experiment Sta- tion, E. Lansing, Mich. Average TOTAL SULPHUR per cent 9.44 9.43 9.43 9.43 9.17 9.16 SULPHID SULPHUR Zine chlorid per cent 9.36 9.39 Cooperative results on THIOSULPHATE SULPHUR a : Zine | Cad: Vee sium, | Todin | chloria | gion | “A percent | percent | per cent | percent | per cent 9.43 9.05 0.07 0.04 0.15 9.32 9.06 0.07 0.04 0.17 = aoe aioe 0.06 za3 aah —— eles 0.06 = ent* 9.38 9.06 0.07 0.04 0.16 pe 8.684 Boies 8.708 aes 8.705 ta ag arr): 8.75 8.45 8.69 ee 8.72 8.45 9.01 | 8.69 8.98 8.71 9.00 8.70 0.16 0.27 ® Weighed as sulphur. lime-sulphur Solution 1. THIOSULPHATE SULPHUR 1920) ROARK: “B” “Cr “sp” per cent per cent per cent 0.08 0.11 0.15 0.08 0.11 0.17 ous 0.11 aoe aL 0.08 ss REPORT ON INSECTICIDES 341 SULPHATE SULPHUR MONO- CALCIUM F ; SULPHUR oxIp Brae aaa Kodi EQUIVALENT (CaO) per cent per cent per cent per cent 0.01 0.02 pate 1.82 3.39 0.01 = us 1.81 3.43 0.02 ees art aoe 4 Baie 0.02 an eee [oF ee 0.02 0.02 cee 1.82 3.41 meas awre ae 1.65 3.47 es fee ee 1.66 3.47 aS aan Eee 1.67 —— a2 ea ae 1.66 3.47 pee ae a 1.75 3.40 =e Bee WS es 1.72 pas peme! ars pak 1.73 3.40 see ate hd 1.73 3.41 ee Eee es, 1.73 3.41 See 0.02 0.02 ee 1.79 3.44 0.02 0.03 ees 1.84 3.43 0.02 0.02 aes 1.82 3.39 open, aut eS 1.85 ee 0.02 0.02 See 1.83 3.42 a cet = 1.85 see == eae eee 1.86 oe ays ne pees tas 1.86 i 0.02 0.02 0.02 1.86 3.50 0.02 0.01 0.02 1.88 3.49 0.02 0.02 0.02 1.87 3.50 eee pepe ANE 1.83 Jae eee 2 aaee 1.83 sae fe, —— ee 1.83 a shou —— aoe 1.84 see — 2 eae 1.82 ae ees aes ae 1.78 nie dee Ue eale 1.83 ee == = eee 1.81 ae ie 1.81 ea 342 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 ANALYST R. N. Miller Averages sass 2.2 W. W. Webber Ayeragelseese 2c 8e3 R. CG. Roark Awverageno nesses General average ANALYST J. J.T. Graham AVGSARG. 24a =sa-5 36 TOTAL SULPHUR per cent 9.09 9.10 SULPHID SULPHUR Cooperative results on THIOSULPHATE SULPHUR . Cad- Z ‘ A chlorid | ium, | Todin per cent | per cent | per cent 8.56 8.67 4.92 8.56 8.51 6.12 awed 8.84 4.88 = 8.65 6.91 = Ne 8.79 aa 8.56 8.69 5.71 Zin Cad ae chlorid | "ium | “A per cent | percent | per cent 0.10 0.06 0.56 0.11 0.06 0.66 use ein 0.79 os Line 0.54 0 oats 0.54 P23 ries 0.48 ee ae 0.38 we aos 0.43 esse ath 0.41 Bees ota 0.31 — Lelie 0.28 =e oe 0.31 0.11 0.06 0.47 Cooperative resulls on TOTAL SULPHUR per cent 9.27 9.24 SULPHID SULPHUR 7: Cad- Zine . P chloria | aium | Todin percent | per cent | per cent 4.85 5.06 4.83 4.81 5.14 4.99 cee ee 4.80 4.83 5.10 4.87 THIOSULPHATE SULPHUR ye Cad- Z * wae chioria | "pba, |) 708 per cent | per cent | per cent 4.07 3.80 4.08 4.07 3.82 4.21 nie ei 4.12 328 es 4.33 aes Bee 4.25 ae ES 3.91 4.07 3.81 4.15 1920] ROARK: REPORT ON INSECTICIDES lime-sulphur Solution 1.—Concluded. 34+ THIOSULPHATE SULPHUR SULPHATE SULPHUR MONO- CALCIUM ; i SULPHUR OxID ane Catnip Todin EQUIVALENT (CaO) per cent per cent per cent per ceni trace trace 0.29 1.67 3.41 ——— Sate 0.37 1.61 3.37 ees HEE 0.36 15 a ae 328 0.35 1.67 pete 4 ==3 = 1.63 aes) aoe dus 2, cud 1.69 ees at. rae pe 1.75 382 Be —_ oe iy? oie = See ee 1.75 a5 ais ates sede 1.76 a =! ae — 1.78 = J Ly ——_ as 1.76 eas trace 0.34 1.69 3.39 0.02 ee 1.35 3.52 0.03 fe 5 1.35 3.54 0.03 ae 1.35 3.53 So 1.72 3,41 see 1.75 oP ss52 1.74 3.41 0.02 1.74 3.44 ' lime-sulphur Solution 2. THIOSULPHATE SULPHUR SULPHATE SULPHUR MONO- CALCIUM : SULPHUR OxID “Br “cr “p” ee Cem Iodin EQUIVALENT (CaO) percent | percent per cenl per cent per cent per cent per cent 4.21 4.10 3.91 0.02 0.02 0.14 0.97 1.84 4.19 4.04 3.93 0.02 0.02 0.14 0.94 1.97 4.19 3.43 os bate aoe 0.31 0.97 ——- == 3.56 sand = ee = 0.36 0.89 See =a 3.58 eles mdi aus 0.36 0.93 aes Bee 3.32 235 st ea 0.25 1.02 fuse ean 3.91 acre i rwe sa acca a f = 4.19 es Bai: wae Biss 55 4.20 3.77 3.92 0.02 0.02 | 0.26 1.91 344 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. JII, No. 3 ANALYST R. M. Chapin Average__-__- D. K. French Average-______ GC. H. Robinson Average-__-__- O. B. Winter Average--_--_- R. N. Miller Average. --__- W. W. Webber Average-_-._-- F. C. Cook Average-_____ S. D. Averitt Average. ----- R. C. Roark Average- - ---- General average ® Weighed as sulphur. TOTAL SULPHUR Cooperative results on SULPHID SULPHUR Zine chlorid per cent | per cent 4.50 4.19 4.17 4.01 chlorid per cent Cad- mium 3.88 3.88 4.86 4.80 4.84 4.97 4.97 4.97 4.65 4.99 4.82 4.8 Iodin per cent THIOSULPBATE SULPHUR Zine chlorid per cent Cad. mium chlorid per cent 3.82 3.82 4.26 4.27 4.29 “An ee le te nee 1920] ROARK: REPORT ON INSECTICIDES lime-sulphur Solution 2.—Concluded. THIOSULPHATE SULPHUR 4.20 4.23 4.22 per cent 3.85 SULPHATE SULPHUR Zinc chlorid per cent Cadmium chlorid per cent Todin per cent MONO- SULPHUR EQUIVALENT CALCIUM OxID (CaO) 3.88 0.02 0.05 0.03 per cent 0.03 0.05 0.05 0.05 0.01 0.01 0.01 4.23 ® Weighed as sulphur. 3.77 3.91 346 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 Cooperative results on SULPHID SULPHUR THIOSULPHATE SULPHUR TOTAL SNM STS SULPHUR| 7; Cad- : Zi eae chlecid),|, age | Todin enact A per cent | per cent | percent | per cent | per cent per cent R. M. Chapin 23°75) ||) ae cee 0.72 0.98 PAG (a) waa 3 uy ia ree BE 0.74 0.98 AVerage so aes PE Ye Aaya | aie | (fora) scope lamer es [an 07 (5) 0.98 J. J. T. Graham 23078 leo) eee) eee: 6 Ooo 0.98 23e70.0|| paaeeO a een) eal) |) O92 0.82 Averages a2) senses 230A | 2 aon) sees, kro. 10:06 0.90 D. K. French 22.73 | 22.51 | 22:78 | 22:60'| 0.88 0.98 DHL | 22 ae | Se 2264. 1093. 0.94 aoe ce 2265. hrseee ESte Average: 22 ase 22.77 | 22.33 | 22.73 | 22.63 | 0.91 0.96 C. H. Robinson 23.90 | 22.62 | 23.21 | 22.80 | 0.85 0.82 23.97 | 22.70 | 23.28 | 22.75 | 0.86 0.83 (Average eeee. oe 23.94 | 22.66 | 23.25 | 22.78 | 0.86 0.83 O. B. Winter te BR Cues BEE‘ aes 0.92 0.92 vere ppt er sae Tas 0.95 0.92 ear pens a) he =e a 0.95 chen Ae 2 age Te! ee 0.92 ssf Es Ser pares bag. ed 0.92 ANeragé-==-. 55=-2)-< 2 AS, 2s Nh eS Oe. 0.93 R. N. Miller 23.78 |) 23:32) ||. 23.64 | --2- 0.74 1.42 23.82) |) Zoo) |) zeae) |) oes 0.77 1.37 ace Mery Ap 2° pie. ‘e: a 1.33 Average: ous 23.80 | 23.39 | 23.40] ---- 0.76 1.36 W. W. Webber Q3:89ri kh Vas 286.0 -— a 0.74 0.60 24:38) ||| 21.67 |) 20:98 |" 222 0.76 0.63 DAQOO*| oes 4, Se oer cae ae 25:44 ||| eee ae aa a 7 2348).|) Seee ae ee wee = Average... 24.44 | 21.59 | 22.42) _--- 0.75 0.62 R. C. Roark 2413") Q2G6NT- fe es 0.67 1.18 2SiGo |) ame ee OES 0.68 1.09 myerage._ == === 23.98 | 22.36) -.-- ee (2 0.68 1.14 General average. __-- 23.89 | 22.46 | 22.98 | 22.42 | 0.82 1920) ROARK: REPORT ON INSECTICIDES 347 lime-sulphur Solution 3. THIOSULPHATE SULPHUR SULPHATE SULPHUR MONO- CALCIUM k . SULPHUR oxID Zine Cadmium EQUIVALENT (CaO) BB Kee “pb” chlorid chlorid Todin per cent per cent per cent per cent per cent per cent per cent Ber alo es =a eh: ae 4.72 er Sse =e ae eee aaa ae 4.69 oe aaa —_ eps ee sess = 4.71 yeas wet ano Paes non a lobe a 4.60 9.73 eel vo a3 ae 2222 pias 4.63 9.48 oe? = se =a3 —- ae 4.62 9.61 Aas 0.06 0.06 ee 4.54 9.19 az* oa a 0.06 0.06 3228 4.59 9.21 aoe ae aces 0.06 0.06 Lhe 4.57 9.20 0.83 =2ah Lik 0.08 0.08 0.11 4.72 9.17 0.84 oe eae 0.09 0.07 0.12 4.71 9.17 0.84 = ==== 0.09 0.08 0.12 4.72 Oily ae mens aoe baie kite et 4.66 pa eo FE ie mee inl rey 4.67 maa ag fare iba we cil aud 4.67 he _ ol ee Bik asia et = 4.55 eee 0.84 oahu wh 0.05 0.05 0.06 | 4.64_ 9.34 348 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 Mr. R. M. Chapin of the Bureau of Animal Industry, sent in the following results on these solutions when analyzed according to his methods!. These methods appear to be desirable in many ways, and the referee regrets that they were not more extensively tested by the members of the association this year. Analyses of association lime-sulphur samples according to Chapin’s methods. DETERMINATION Oca ahaa | park per cent per cent per cenl Hydrogen sulphid sulphur (reaction figure) - -- —___ 0.12 0.05 0.06 0.12 0.05 0.06 Polysulphidsulphur:=—— ose eee ee 7.12 3.66 18.28 (5 3.68 18.34 Thiosulphate sulphur22- ee82 oe eee ee 0.09 4.20 0.84 0.09 4.20 0.85 Sulphid acid figure (monosulphid sulphur) ___ ~~ ~~~ 2.00 1.04 4.82 2.00 1.04 4.81 Sulphid sulphur (sum of monosulphid and poly- sulphid‘slphur) 2-2-2 =32 sees eee 9.14 4.71 23.13 COMMENTS BY ANALYSTS ON LIME-SULPHUR METHODS. S. D. Averitt—The methods as given this year for total sulphur and total lime required too much manipulation and time for this character of work. Moreover, figures for total sulphur and lime by these methods are practically identical with those obtained by shorter methods recommended in 1913. a DETERMINATION 1916 InsTRUCTIONS | 1913 INSTRUCTIONS + | 5 Totalsulphurss_- oa - = 25s oe ee 8.90 8.88 { 8.92 8.89 y § Totalilime:(CaQ)\4-- 5 sass. 2e ee 3.41 3.39 3.40 3.38 It is impossible to explain the high figures for thiosulphate in Sample 1 by the iodin titration method unless it was made by diluting a lime-sulphur solution with lime water. The low figures for thiosulphate in this sample by the zinc chlorid method are due to adsorption of thiosulphate by the precipitated zinc sulphid. In Sample 2, containing a relatively small amount of sulphid sulphur, adsorption was small; consequently, the figures for thiosulphate by the zine chlorid method are accurate. R. M. Chapin.—(1) In the estimation of total sulphur I dissolved the sodium peroxid — in 50-75 ce. of cold distilled water (best ice-cold) and then pipetted in the sample. I : prefer to deliver all samples in this way? and a clean pipette should be used each time. — 1J. Ind. Eng. Chem., 1916, 8: 151, 339. *Tbid., 151. 1920] ROARK: REPORT ON INSECTICIDES 349 (2) For thiosulphate sulphur I regard the cadmium chlorid method as treacherous. If carried through cold it seems to give low results, particularly by the referee’s modi- fication, while if heat is applied before filtration the results appear erratically high. Zinc chlorid, giving more nearly uniform results under varied manipulation, is more dependable. My modification gives higher results than the referee’s method, and since the difference in the titrations varies in proportion to the amount of thicsulphate present, I conclude that adsorption is here a factor to be taken into consideration. (3) Regarding the so-called ‘‘iodin titration methods” originated by Harris, I see no reason to change my previous opinion, namely, that they are to a significant degree unsound and therefore untrustworthy for exact work. (4) In respect to my own methods, it may be true that their rationale is such as normally to produce slightly high results. But it is certainly true that the gravimetric methods normally yield somewhat low results. I have scrutinized each step of my processes and can find no cause which, given proper execution, will lead to any significant error; consequently, I can only believe that my methods are at least as accurate as the ‘gravimetric methods. The true values very likely lie in between. D. K. French—In virtually all cases four independent samples were taken from each solution, made to volume, and the work done upon these different samples. Each figure reported represents the average of five or more determinations made upon the same sample. In other words, on the first sample from Solution 1 the titration was run on two or three 10 cc. portions, one or two 20 cc. portions, possibly a 50 cc. portion, and possibly a 5 cc. portion, the results calculated in all cases and the average reported. When only one or two figures are given, the results on the other samples were so far off and knowledge of the reason why they were so far off was so sufficiently clear that the results were not tabulated. Our laboratory being located in an office building, we are slightly handicapped for want of steam bath conditions, having a very well developed hot plate and sand bath system. This, however, in connection with some of the sulphur figures was not satisfactory, and a certain amount of oxidation took place in the drying. Where any indication of this was noted the results of the determinations were disre- garded, as in almost every case they were well off. Regarding the choice as to methods, the men doing the work are very much in favor of the iodin method, feeling that the results are obtained more rapidly that way. C. H. Robinson —Results throughout for sulphid and thiosulphid sulphur by iodin and zinc methods were closely concordant. Cadmium methods invariably gave higher sulphid sulphur and lower thiosulphate sulphur. The iodin methods are more con- venient for thiosulphate sulphur, but, if anything, less so for sulphid sulphur. More time is consumed in washing and dissolving sulphur than in washing zinc sulphid. Difficulty was experienced in getting uniform results for calcium oxid by the hydro- chloric acid method with Sample 2. The oxidation method gave higher but closely concordant results. J. J.T. Graham.—Of the methods for thiosulphate sulphur in lime-sulphur, I much prefer the zinc chlorid method, as it has invariably given good results. The cadmium chlorid methods do not give as uniform results in my hands. In the iodin titration method great difficulty was experienced in determining the end point in the mono- sulphur equivalent titration, as the color of the sodium nitroprussid was discharged very slowly and considerable excess of iodin may be added here. After making a num- ber of preliminary titrations I made the ones reported with extreme care and believe they are as accurate as the method will give. I had no difficulty whatever in deter- mining the end point in the final titration. As the value for thiosulphate sulphur is obtained from the difference between the monosulphur equivalent titration and the total titration, any error in the former will affect the thiosulphate values. In Method 350 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 B for thiosulphate sulphur the end points are very easily determined and this method gave results which agreed very closely. Method C for thiosulphate sulphur was very unsatisfactory. R. N. Miller.—Of the various methods submitted the zinc chlorid method seems to be the least undesirable. In it the chief objections are to be found in the method of making the precipitation, and in handling the filter paper after the precipitation. The manipulation as carried out by Mr. W. J. Morgan of the Insecticide and Fungicide Laboratory of the Bureau of Chemistry, greatly lessens the difficulty. His method is as follows: Pipette the aliquot into the beaker and add the precipitant without dilution. Move the beaker in a circular movement without slopping on the sides of the beaker. Rinse the sides once and throw immediately on a filter paper. Wash with water. Place the paper opened out on the sides of the beaker in which the precipitation was carried out and rinse the precipitate into the beaker. Add the peroxid, leaving the paper on the sides of the beaker. When the precipitate is all oxidized drop the paper into it and let stand for a little while. Then acidify and filter off the paper and continue as in the method furnished by the referee. When carried out in this way the filtration is as fast as the cadmium chlorid method. The objection to the cadmium chlorid method that makes it less desirable than the zinc chlorid method is the difficulty experienced in oxidizing the sulphid pre- cipitate. This takes two or three times as much peroxid and considerably longer than the zinc sulphid. Aside from this defect the method seems to be about as good as the zinc chlorid method. In the iodin titration method the great variation in the results would indicate that even if the method were scientifically correct, it could not be carried out with accuracy except by one skilled in the manipulation of the method. The end point for the monosulphur equivalent sulphur is not definite without the nitroprussid indicator and the addition of the reagent does not increase the definiteness. If added too soon, the color can be shaken out in 30 seconds. The only way to tell when to add it is to watch the fading of the yellow color, and when the right point is reached there is no further need for an indicator. This point can not be told with definiteness. O. B. Winler—Sample 1. This sample exhibited characteristics entirely different from any samples of lime-sulphur solution, either commercial or homemade, that have ever been experienced in the laboratory of the Michigan Agricultural Experiment Sta- tion, and some difficulty was found in arriving at the end point for the monosulphur equivalent, especially when using nitroprussid of sodium. The results for thiosulphate by the three methods do not agree satisfactorily. Neither do the results by the iodin or zine chlorid methods agree so well as they should. This variation in the results is believed to be due to some peculiarity in the sample. Two separate solutions of Sample 1 were prepared, one portion being analyzed immediately and the other 12 hours after preparation. The second solution was kept in a well-stoppered flask and remained perfectly clear until analyzed. Samples 2 and 3. The results on both samples by all three methods are practically the same, the greatest variation in the averages of all the results being 0.06 percent. The difference between the maximum and minimum results by the iodin and zine chlorid methods on these two samples is 0.08 per cent. Comments: (1) In using sodium nitroprussid as an indicator for the end point in the monosulphur titration by the iodin method, it was found necessary to add the iodin very slowly, drop by drop, stirring vigorously after each addition, as the blue color is discharged very slowly. There seems to be no special advantage in using the nitro- 1920) ROARK: REPORT ON INSECTICIDES 351 prussid, for, if one is careful, the end point can be determined very accurately without the indicator. (2) The results by the cadmium chlorid method were not, generally, so satisfactory as by the two other methods. On Sample 1 the results by this method were higher and on Samples 2 and 3 lower than by the iodin or zinc chlorid method. The dif- ferences, however, were not great in either case. No explanation is offered for the high results, but the lower results may possibly be explained by occlusion. The cadmium sulphid precipitate is much more flocculent than the zinc chlorid precipitate. It is suggested that an approximately N/10 solution of cadmium chlorid be used and that the solution be stirred vigorously during the addition of cadmium chlorid. (3) In point of accuracy there is very little difference between the three methods, a fact that has been well substantiated by previous collaborative work and by a large amount of unpublished work performed in the laboratory of the Michigan Agricul- tural Experiment Station. The iodin method has many advantages over the others, especially in the time required to complete an analysis, and on this account, without sacrificing accuracy, is preferable. R. C. Roark.—The use of sodium nitroprussid enables the analyst to get a more accurate and definite end point in the titration for monosulphur equivalent, but 0.1—0.2 ec. of N/20 iodin solution is as close as either the monosulphur equivalent or thio- sulphate titration can be made according to the Harris-Averitt method, even with the use of sodium nitroprussid and starch paste to indicate the end points. I regard the method of determining sulphid sulphur by weighing the sulphur precipi- tated from solution by iodin as worthless, and if it is oxidized to sulphate and weighed as barium sulphate the method is more tedious and less accurate than the zinc chlorid method!. I do not believe that the analysis of a lime-sulphur solution by the zinc chlorid method will consume any more time (exclusive of evaporations, which may be done overnight) than when made according to the iodin titration method. The only determination common to the two methods which can be made in less time by the iodin titration method than by the zinc chlorid is that of thiosulphate sulphur, and this saving of time is offset by the extended period through which the precipitated sulphur in the determina- tion of sulphid sulphur must be allowed to stand before it can be filtered with any approach to accuracy. The method for total sulphur has been criticized because of unnecessary manipula- tion. To obtain accurate results it is necessary to remove silica from a solution before determining sulphates. The method recommended by the referee, which is the pro- cedure of Johnston and Adams?, has been adopted by the association for the determina- tion of large amounts of sulphate in water’. In regard to the determination of lime, it is unnecessary to go through the steps for the removal of iron and aluminium, as they are present only in mere traces, if at all, in lime-sulphur solution. In regard to the cadmium chlorid method, the work this year shows that it possesses some disadvantages and no advantages not possessed by the zinc chlorid method. ~ Inconclusion, I am strongly of the opinion that the zine chlorid methods, as presented to the association this year, together with certain precautions noted in this report, are the most accurate ones on which any cooperative work has been done, and should be adopted as official by the association. 1 J. Assoc. Official Agr. Chemists, 1915, 1: 76. 2.J. Am. Chem. Soe., 1911, 33: 844. * Assoc. Official Agr. Chemists, Methods, 1916, 44. 352 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 PREPARATION OF AMMONIACAL ZING CHLORID SOLUTION. The preparation of ammoniacal zinc chlorid has been studied by many including: Sutton’, Richardson and Aykroyd?, McDonnell’, Blockey and Mehd?, Proctor®, McCandlish and Wilson®, and Bennett’. From a consideration of the literature and the reports of collaborators, it is recommended that the method of preparation of the ammoniacal zine chlorid solution include a statement of the amount of ammonia to be added. To test the effect of zinc chlorid solutions made up in different ways upon the determinations of sulphid and thiosulphate sulphur, the lime- sulphur solutions sent out for cooperative study were analyzed by the zinc chlorid method, using the following zinc chlorid solutions: (a) A concentrated solution prepared by dissolving 225 grams of zine chlorid in water, adding 465 cc. of ammonia water (sp. gr. 0.90) and diluting to 1 liter; (b) the solution recommended in this year’s instructions, con- taining 50 grams of zinc chlorid, 50 grams of ammonium chlorid and 120 ce. of ammonia water (sp. gr. 0.90) per liter; and (c) a N/10 solution prepared essentially as directed by Bennett’, containing 6.8 grams of zine chlorid, 50 grams of ammonium chlorid and 25 ce. of ammonia water (sp. gr. 0.90) per liter. Results are as follows: Comparative results for sulphid and thiosulphate sulphur in A. O. A. C. lime-sulphur samples, using different zine chlorid solutions. SULPHID SULPHUR THIOSULPHATE SULPHUR LIME-SULPHUR (a) (b) (c) (a) (b) (c) per cenl per cent per cent per cent per cent per cent Solution 1___ 8.61 8.67 8.82 0.10 0.09 0.10 Solution 2_ _- 4.44 4.70 4.61 3.97 4.09 4.14 Solution 3__-]| 22.47 22.36 aoe. 0.68 0.68 eae From these results it is seen that the figures obtained with different zine chlorid solutions do not differ more than duplicate determinations made with the same solution often do. ‘Sutton. Volumetric Analysis. 10th ed., 1911, 342. 2 J. Soc. Chem. Ind., 1896, 15: 171. 3U.S. Bur. Chem. Bull. 152: 70. 4 J. Soc. Chem. Ind., 1912, 31: 369; J. Am. Leather Chem. Assoc., 1914, 9: 176. 5 Leather Industries Laboratory Book on Analytical and Experimental Methods. 2d ed., 1908, 55. 6 J. Am. Leather Chem. Assoc., 1913, 8: 28; 1914, 9: 203, 205. 7 [bid., 1916, 11: 110, 112. 8 Jbid., 1916, 11: 112. j 4 1920] ROARK: REPORT ON INSECTICIDES 353 DISCUSSION. In considering a method for the analysis of a commercial product the first question to consider is—are any substances present that will inter- fere with the method and vitiate the results? This is particularly im- portant when the material under examination is a solution, as many substances exist in solution that have never been isolated in the pure form. Hence, in considering different methods for the analysis of lime- sulphur solution it is first necessary to take into account all the different compounds that can exist in such a solution prepared by any of the methods ordinarily employed. RECENT LITERATURE ON LIME-SULPHUR SOLUTIONS. For those who may be interested the following references to articles on the composition or analysis of lime-sulphur solutions are given: Averitt!, Roark?, Thompson and Whittier’, Ramsay‘, Auld®, Van Slyke'*, Tartar’, Eyre and Salmon’, Oberfell®, Len!°, Proctor!!, Chapin!?, Green}, and Blumenthal and Averitt! AMENDED WORDING OF LIME-SULPHUR METHODS. The following methods are exactly the same in principle as those previously printed'®. Certain precautions have been added, however, and the descriptions made fuller, so that it is believed the methods here given will be easier to follow. LIME-SULPHUR SOLUTIONS. PREPARATION OF SAMPLE. Weigh 10 grams of the solution in a weighing pipette, transfer to a 250 cc. graduated flask, and immediately dilute to the mark with recently boiled and cooled water. Mix thoroughly and transfer to a number of small bottles, entirely filling them, and avoiding contact of the solution with air as much as possible. Stopper the bottles, seal with paraffin and preserve in a dark, cool place. 1 J. Ind. Eng. Chem., 1916, 8: 624; J. Assoc. Official Agr. Chemists, 1915, 1: 74- 2J. Assoc. Official Agr. Chemists, 1915, 1: 76. 2 Del. Agr. Expt. Sta. Bull. 105: 5, 29. ‘J. Agr. Sci., 1914, 6: (1), 194. 8 J. Agr. Sci., 1915-16, a “av), “473. 9 J. Am. Leather Chem Assoc., 1915, 10: 253. 10 J. Soc. Dyers Colourists, 1914, 30: 277. 1 Leather Industries Laboratory Book on Analytical and Experimental Methods. 2d ed., 1908, 56. 12, J. Ind. Eng. Chem., 1916, 8: 153. ted Union of South Africa, Dept. of Agr., 3rd and 4th Reports of the Director of Veterinary Research, ov. 1915, p. 175. 4J. Am. Chem. Soc., 1916, 38: 1701. 18 Assoc. Official Agr. Chemists, Methods, 1916, 76. 354 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 3 TOTAL SULPHUR.—OFFICIAL. DETERMINATION. Dissolve 2 or 3 grams of sodium peroxid in 50 cc. of cold water in a 250 cc. beaker. Transfer a 10 cc. aliquot of the solution prepared for analysis as directed above to this aqueous solution of sodium peroxid, constantly keeping the tip of the pipette just under the surface of the liquid until necessary to raise it for drainage at the end. Use a clean, dry pipette for measuring each portion!. Coyer the beaker with a watch glass and heat on the steam bath, with occasional stirring, until all the sulphur is oxidized to sulphate, which is indicated by the disappearance of the yellow color. Wash the watch glass and the sides of the beaker, acidify with hydrochloric acid, evaporate to complete dryness, treat with water acidified with hydrochloric acid, boil, and filter to remove silica. Dilute the filtrate to 300 cc., add 50 cc. of concentrated hydrochloric acid*, heat to boiling, and precipitate with 10% barium chlorid solution slowly and stirring constantly. This should be added at such a rate that about 4 minutes are required in running in the amount necessary (11 cc. for 1 gram of barium sulphate). The rate is best regulated by attaching a suitable capillary tip to the burette contain- ing the barium chlorid solution. Evaporate to dryness on the steam bath, take up with hot water, filter through a quantitative filter paper, wash until free from chlorids, ignite carefully and heat to constant weight over a Bunsen burner. Calculate the sulphur from the weight of barium sulphate, using the factor 0.13734. Sulphur-free reagents, which can easily be procured, should be used for all work. SULPHID SULPHUR.—OFFICIAL. REAGENT. Ammoniacal zine chlorid solution—Dissolve 50 grams of pure zinc chlorid in about 500 cc. of water, add 125 cc. of ammonia water (sp. gr. 0.90) and 50 grams of ammonium chlorid and dilute to 1 liter. (See p. 352.) DETERMINATION. To 10 or 15 cc. of water in a small beaker add, as directed under total sulphur, a 10 cc. aliquot of the solution prepared for analysis. Calculate the amount of ammonia- cal zinc chlorid solution necessary to precipitate all the sulphur in the aliquot, and add a slight excess. Stir thoroughly, then filter, wash the precipitate 3 or 4 times with cold water, and transfer filter paper and precipitate to the beaker in which the pre- cipitation was made. Cover with water, disintegrate with a glass rod and add about 3 grams of sodium peroxid, keeping the beaker well covered with a watch glass. Warm on the steam bath with frequent shaking until all the sulphur is oxidized to sulphate, adding more sodium peroxid if necessary. Make slightly acid with hydrochloric acid, filter to remove shreds of filter paper, wash thoroughly with hot water, and determine the sulphur in the filtrate exactly as under VII, 69°. THIOSULPHATE SULPHUR.—OFFICIAL. To 500 cc. of water in a 200 cc. graduated flask, add, as directed under VII, 69°, 50 ce. of the solution prepared for analysis. Add a slight excess of the ammoniacal zine chlorid solution and dilute to the mark. Shake thoroughly and filter through a dry filter. To 100 cc. of the filtrate add a few drops of methyl orange or methyl red and exactly neutralize with N/10 hydrochloric acid. Titrate the neutral solution with approximately N/20 iodin solution, VII, 3 (c)*, using a few drops of starch solution 1 J. Ind. Eng. Chem., 1916, 8: 152. 2 J. Am. Chem. Soc., 1911, 33: 844. 3 Assoc. Official Agr. Chemists, Methods, 1916, 76. * Thid., 63. ou 1920} ROARK: REPORT ON INSECTICIDES as indicator. From the number of cc. of iodin solution used, calculate the thiosulphate sulphur present, using the factor AsO, 1.2959 =thiosulphate sulphur (8). TOTAL LIME.—OFFICIAL. To 25 ce. of the solution, prepared as directed under VII, 68', add 10 cc. of con- centrated hydrochloric acid, evaporate to dryness on the steam bath, treat with water and a little hydrochloric acid, warm until all the calcium chlorid is dissolved, and filter from sulphur and any silica that may be present. Dilute the filtrate to a bulk of 200-250 cc., heat to boiling, add a few excess cc. of ammonia, and then an excess of a saturated solution of ammonium oxalate. Continue the boiling until the precipitated calcium oxalate assumes a well-defined granular form, allow to stand for an hour, filter, and wash a few times with hot water. Ignite in a platinum crucible over a blast lamp to constant weight, and weigh as the oxid!. SUGGESTIONS FOR FUTURE WORK. (1) The distillation method is not applicable to London purple with- out previous destruction of the organic matter present, otherwise the distillate is colored purplish or reddish by the dyes that are carried over in the distillation stream. If the organic matter is destroyed by heating with nitric and sulphuric acids it is easier to proceed with the reduction with potassium iodid and sulphuric acid and determine the arsenic’. The referee has tried heating London purple with a mixture of zinc oxid and sodium carbonate in a muffle for the destruction of organic matter, and has found this very satisfactory, no arsenic being lost by volatilization. The sample of London purple used in 1910 for the asso- ciation work was treated in this way and the total arsenic then deter- mined by the official distillation method. The total arsenic found, calculated as As.O3, was 38.15 per cent. The total arsenic found by the present official methods* was 38.22 per cent, this being the sum of the average As.O; and As.Q;, calculated as As.O;, found by the cooperators in 19104. The mixture of zinc oxid and sodium carbonate, which is prepared by thoroughly mixing 4 parts of zinc oxid with 1 of dry sodium carbonate, was proposed by Ebaugh and Sprague® as a fusion mixture for ores con- taining arsenic. It has been used successfully by Krickhaus® and Low’. In analyzing a London purple by this method, the sample should be thoroughly mixed with the zinc oxid-sodium carbonate mixture, and then a layer of the latter put on top, the whole being contained in a shallow porcelain crucible. Place the crucible, uncovered, in the elec- tric muffle and heat gradually, finally for about 15 minutes with full 1U.S. Geo. Surv. Bull. 422: 230. ? Assoc. Official Agr. Chemists, Methods, 1916, 66. 3 Tbid., 17 and 19. 4U.S. Bur. Chem. Bull. 137: 39. 5 J. Am. Chem. Soc., 1907, 29: 1475. Et Mining J., 1910, 90: 357. A. H. Low. Technical Methods of Ore Analysis. 5th ed., 1911, 47. 356 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 heat. This treatment will completely burn off all organic matter with- out driving off any arsenic. The mass will not fuse with this short heating, and, in fact, should not be allowed even to sinter. After all organic matter is destroyed, transfer to a distillation flask and proceed as usual!. (2) The referee has tested the method of Gyory? for the titration of arsenic trioxid in hydrochloric acid solution with potassium bromate solution and has obtained excellent results. It is believed that this method should be adopted as an optional official method for the titration of arsenic in the distillate after distillation as arsenic trichlorid. As an example of the results yielded by this method, the referee obtained 38.24 per cent in the 1910 association sample of London purple as against 38.22 per cent by the present official method and 38.15 per cent by the iodimetric titration of the distillate. This method has been used very successfully by Nissenson and Siedler’, Rowell‘, and Schmidt? for the determination of antimony. Low® includes this method in his book. Jannasch and Seidel? used this method for titrating the arsenic in the distillate in the distillation of arsenic as arsenic trichlorid. They found that the titration was independent of the volume and of the degree of acidity between 10 and 40 per cent. The best results were obtained when from 10 to 25 per cent hydrochloric acid (sp. gr. 1.19) was present. Two or three drops of a 0.1 per cent aqueous solution of methyl orange serve as indicator, the end point being shown by a change from pink to colorless. The advantages of this method over the iodimetric for determining arsenic in a distillate containing arsenic trichlorid are: (1) No neu- tralization of the distillate is required. This process takes considerable time and consumes large amounts of alkali. Furthermore, we have found in this laboratory samples of sodium hydroxid, purified from alcohol, that contained appreciable amounts of iodin-consuming sub- stances, necessitating the rejection of the alkali for this purpose. (2) The methyl orange solution is stable, whereas starch solution must be freshly prepared to secure good results. (3) The method is cheaper as regards reagents, no alkali or sodium bicarbonate being used, and potassium bromate being cheaper than the equivalent amounts of iodin and potas- sium iodid used in preparing standard iodin solution. 1 Assoc. Official Agr. Chemists, Methods, 1916, 63, 4. 2 Z. anal. Chem., 1893, 32: 415. 3 Chem. Zilg., 1903, 27: 749. 4 J. Soc. Chem. Ind., 1906, 25: 1181. * Chem. Zig., 1910, 34: 453. 6A. H. Low. Technical Methods of Ore Analysis. 5th ed., 1911, 34. 7 J. prakt. Chem., 1915, 91: 133. 1920] ROARK: REPORT ON INSECTICIDES 357 Titration of As™ in hydrochloric acid solution with potassium per- manganate has been tested by Moser and Perjatel', who recommend that the titration be made in the cold, drop by drop, in a solution con- taining only a slight excess of acid. In the hands of the referee this method was found to be too slow, and to yield erratic results due to indefiniteness in the end point, even when manganese sulphate (see Zim- merman-Reinhardt method for iron?) was added to the solution. (3) Some method should be devised for removing the color from arsenic solutions without loss of arsenic, either by adsorption or oxida- tion of As.O; to AssO;. The referee has tested talcum, U. S. P., and animal charcoal, U. 8. P., on acid solutions of London purple, but with- out effect. The following are suggested as worthy of trial: Blood char- coal, fuller’s earth, kaolin and kieselguhr. Blood charcoal was tried by Chapin®, who found that it oxidized arsenites to arsenates. From fuller’s earth Lloyd’s reagent, or hydrous aluminium silicate, can be prepared, which would probably decolorize a solution very effectively. RECOMMENDATIONS. It is recommended— (1) That the method for the determination of AsoO; in lead arsenate as described on p. 332 be adopted as a tentative method. (2) That the method for the determination of As.O; in lead arsenate as described on pp. 363-4 be adopted as a tentative method. The method as there given is the same as presented on pp. 332-4, except that a pre- liminary evaporation with hydrochloric acid is called for, this being necessary to destroy nitrates and peroxids which are sometimes found in commercial samples. (83) That further study be made of methods for the determination of copper, lead and zinc in such preparations as Bordeaux-lead arsenate, Bordeaux-zinc arsenite, ete. (4) That the so-called zinc chlorid methods for the analysis of lime- sulphur solution be adopted as official, using the amended wording given on pp. 353-5. (5) That cooperative work be done on the Gyory method for titrating As™ in hydrochloric acid solution with a solution of potassium bromate. (6) That Method I for total arsenic oxid‘ be dropped. (This method is not published in the Association of Official Agricultural Chemists, Methods, 1916.) 1 Monatsh., 1912, 33: 751. ?A.H. Low. Technical Methods of Ore Analysis. 5th ed., 1911, 131. * J. Ind. Eng. Chem., 1914, 6: 1002. ‘U.S. Bur. Chem. Bull. 107, rey.: 28. 358 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 (7) That the methods for the determination of moisture, free acetic acid and free ammonia proposed in 1910! and adopted as official, final action in 1912*, be dropped. (These methods are not published in the Association of Official Agricultural Chemists, Methods, 1916.) The following recommendation, which was made in 1915, is renewed for final action: (8) That all other methods for insecticides and fungicides be adopted as tentative and official as given in the Association of Official Agricultural Chemists, Methods, 1916, VII, 63-77, except as further modified in this report (see lime-sulphur methods, pp. 353-5). THE OCCURRENCE AND DETERMINATION OF As™ AND As’ IN THE PRESENCE OF EACH OTHER IN ARSENICAL INSECTICIDES. By R. C. Roarx* (Bureau of Chemistry, Washington, D. C.). It has been customary, except in the case of London purple, to regard the arsenic in arsenical insecticides as existing either as an arsenate or an arsenite, rather than as a mixture of the two. It has been found, however, from the results of many analyses made in the Insecticide and Fungicide Laboratory, that this assumption is not correct. Even in the cases of lead arsenate and zinc arsenite, in which the total arsenic is commonly reported as arsenic oxid, As,O;, and arsenious oxid, As:O3, respectively, arsenic has been found in both forms of oxidation. DIRECT DETERMINATION OF As” IN LEAD ARSENATE. At the 1915 meeting of the association the author presented a method for estimating As,OQ; in lead arsenate, but the results obtained by the cooperators by this method were low. This method directed that the sample be boiled a few minutes with dilute sulphuric acid (3 to 4 ce. of concentrated acid to 100 cc. of water), and it was thought that the low results were due to lack of sufficient acid or to insufficient boiling. The effect of concentration of acid is shown in the following table of results on the 1915 association sample of lead arsenate containing lead arsenite: 1U.S. Bur. Chem. Bull. 137: 38. 2 [bid., 162: 49. 3 Present address, General Chemical Co., Baltimore Works, Baltimore, Md. 1920| ROARK: ARSENICAL INSECTICIDES 359 TABLE 1. Effect of acid concentration and time of boiling upon the delermination of arsenic triorid in a mizture of arsenate and arsenite of lead. WEIGHT OF SAMPLE ee WATER TIME BOILED So Aare grams cc. cc. minutes per cent 2.0 5 75 30 | 5.80 Qf, 10 75 30 7.40 2.0 15 75 30 7.75 Del 15 90 40 U5 2.7 15 75 | 40 7.74 2.7 15 60 40 7.70 | The amount of AsO; in this sample, as determined by analysis of the ingredients from which it was prepared, is 7.75 per cent. Additional tests on the time of boiling showed that 30 minutes were sufficient, the time being taken when the solution actually began to boil. As a result of these tests the directions given in the 1915 report on insecticides were changed to the following: TOTAL ARSENIC TRIOXID IN LEAD ARSENATE. Weigh an amount of the powdered sample equal to the amount of arsenic trioxid to which 1000 cc. of the iodin solution are equivalent. Transfer to a 200 cc. graduated flask, add 100 cc. of dilute sulphuric acid (water, 85 cc.; concentrated sulphuric acid, 15 ce.), and boil for 30 minutes. Cool, make to volume, shake thoroughly, filter through a dry filter, take 100 cc. of the filtrate, nearly neutralize with a saturated solu- tion of sodium or potassium hydroxid, using phenolphthalein as indicator (if the neutral point is passed make acid with sulphuric acid again), complete the neutralization with sodium bicarbonate, add 4 or 5 grams in excess, and titrate with N/20 iodin solution in the usual way. The number of cc. of iodin solution used in this titration multiplied by 0.2 gives the per cent of arsenic trioxid in the sample. This method is essentially the same as that used by Haywood and McDonnell' in their investigation of lead arsenates found on the market, except that a greater volume of acid is used and the time of boiling is shortened. The results obtained on 100 commercial lead arsenates by this method are shown in Table 5. Thirty-eight samples contained no arsenic tri- oxid, twenty-seven had less than 0.10 per cent, fourteen had from 0.10 to 0.19 per cent, six contained from 0.20 to 0.29 per cent, five contained from 0.30 to 0.49 per cent, eight contained from 0.50 to 0.99 per cent, while two contained over 1 per cent. The ordinary form of lead arsenite is Pb;(AsO;)2. Assuming all the As.O; present to be combined in this form, it is seen that a number of U.S. Bur. Chem. Bull. 131: 8. 360 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 3 commercial lead arsenates contain appreciable quantities of lead arsenite, for the amount of lead arsenite would be about 4.5 times the arsenic trioxid present. Of fifty commercial lead arsenates examined, Haywood and McDon- nell! found only two to contain more than a trace of As»O3, and one of these samples was labeled lead arsenite. This sample contained 16.31 per cent of As.O;, and 10.01 per cent of As.03;; while the other sample contained only 3.91 per cent of As.O;, and 40.44 per cent of As2Os, all results being calculated on the dry basis. The water-soluble arsenic of both of these samples was high, running 5.56 and 3.28 per cent, respec- tively, calculated as As.O;. This would indicate that all the As.O3 in these samples was not present as lead arsenite, but perhaps as sodium or some other soluble arsenite, for the water-soluble arsenic of Pb3(AsOs3)2 (determined according to the tentative association method for Paris green) is only 0.23 per cent. Of eleven commercial samples of sodium arsenate examined, Haywood and McDonnell? found only traces of AsO; in two or three. This would indicate that As,O; is more likely to be present in a lead arsenate made by some process other than where lead nitrate or acetate is precipitated with sodium arsenate. This method is not applicable if much nitrate is present, as the nitric acid liberated oxidizes some of the As2O; to AseO;, in which form it is not titrated with the iodin solution. To test the action of nitrates, sodium nitrate was added to the 1915 association sample of lead arsenate with lead arsenite in amounts equivalent to 5 and 10 per cent of the whole. Results are as follows: TABLE 2. Effect of nitrates upon the determination of arsenic triorid in a mixture of arsenate and arsenite of lead. coor nimane, | oncemre, | waren | rmenousn | SSRIS EROXD per cent ce. ce. minules per cent 0.00 15 85 30 7.79 7.73 5.00 15 85 30 7.35 7.45 10.00 15 85 30 6.71 6.53 These results show that the method is not applicable in the presence of appreciable quantities of nitrate, but we have found very few com- 1U.S. Bur. Chem. Bull. 131: 9. 2 [hid., 16. 1920| ROARK: ARSENICAL INSECTICIDES 361 mercial lead arsenates to contain more than a trace of nitrate. Haywood and McDonnell! found nineteen out of fifty commercial lead arsenates to contain nitrates in small amounts. DIRECT DETERMINATION OF As” IN ZING ARSENITE. For determining AsO; in zinc arsenites the method presented to the association in 1915 for the determination of As,O; only in lead arsenate was used. As shown later in this paper, this method is not accurate in the presence of nitrates or peroxids, but they are not present in commer- cial zinc arsenites. Iron salts are always present, but only in traces. The C. M. Smith method, modified? was used in determining the arsenic trioxid, and the total arsenic was determined by dissolving the sample in acetic acid, precipitating zinc as the oxalate, and determining arsenic in the filtrate after reduction with potassium iodid and sulphuric acid according to the method of Gooch and Browning’, except that sodium thiosulphate was used, as suggested by Haywood’, to discharge the iodin remaining in solution instead of sulphurous acid. This method was pre- sented to the association in 1915, but was recommended to be discarded because antimony is determined by it in addition to arsenic and only the total arsenic was desired. Each of these methods will determine antimony, so that the results are comparable whether or not antimony is present, which we know is the case in certain samples. TABLE 3. Results on commercial zinc arsenites (dried samples). As:03+ ae ORIGINAL LABORA- A Se A meee pices c Are pete ft Tune | 'sampze | umpen | onty | oNty. | CALCULATED | CALCULATED 7 per cent per cent per cent per cent per cent A Powder 18985 1.75 40.40 41.75 41.91 —0.16 A Paste 22308 1.16 40.15 41.15 41.15 +0.00 A Powder 22973 1.29 41.10 42.28 42.21 | +0.07 A Powder 23034 1.63 40.43 41.85 41.83 +0.02 A Paste 24949 0.32 42.80 43.20 43.08 +0.12 A Paste 24953 2.05 40.70 42.30 42.46 —0.16 A Powder 25188 0.72 40.81 | 41.50 41.43 +0.07 B Powder 16151 0.90 36.90 37.23 37.67 — 0.44 B Powder 16586 1.31 36.23 37.43 37.36 +0.07 Cc Powder 19649 1.98 40.20 41.53 41.90 —0.37 The results are presented in Table 3. It is seen that all commercial zinc arsenites contain some arsenic in the pentavalent form and in rela- 1U.S. Bur. Chem. Bull. 131: 9. 2 Assoc. Official Agr. Chemists, Methods, 1916, 7. 3 Am. J. Sci., 1890, 3rd ser., 40: 66. U.S. Bur. Chem. Bull. 105: 167. 362 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 tively large amounts in some cases. The sum of the separate determina- tions of AsO; and As.O:, calculated to a common basis of As:O3, checks the total arsenic, also calculated as AssO, in nearly all the samples. THIOSULPHATE METHOD FOR THE DETERMINATION OF As’, AND ITS APPLICABILITY TO COMMERCIAL LEAD ARSENATES. This method, which is based on the reaction As.O; +4HI = AsoO3-+ 2H,O+ I, which takes place in acid solution, and the titration of the liberated iodin with standard thiosulphate solution, was first studied by Naylor’. He found that the concentration of the hydriodic acid was important, the greatest reduction being effected with solutions containing not less than 20 per cent of hydriodic acid. Phosphates appeared to retard the reduction, and ferric chlorid affected both the hydriodic acid and the thiosulphate solution used in titrating the liberated iodin. Naylor’s procedure in quantitative work was as follows: An amount of arsenate equivalent to 0.03—0.05 gram of arsenic acid was dissolved in just sufficient water or dilute hydrochloric acid, 5 cc. of 20% hydriodic acid were added, and the iodin titrated with N/10 thiosulphate solution in an atmosphere of carbon dioxid. Fifteen minutes were allowed after decoloration of the solution before taking the final burette reading. The solution of hydriodic acid was prepared by dissolving 34 grains of potassium iodid in 100 cc. of water and adding 25 cc. of hydro- chloric acid (sp. gr. 1.16). As a qualitative test, this method would detect 0.1 mg. of AssO; in the presence of 1 gram of As,O3. Williamson? found that the reduction is complete only in strongly acid solution, and that if such a solution be diluted the reverse action takes place to a certain extent. Sulphuric acid could be used instead of hydrochloric. Williamson called attention to the error caused by the presence of nitrates, which is quite large when the sample is heated with potassium iodid and hydrochloric acid, but is not appreciable if the determination is carried through in the cold. This method has been studied also by the following: Gooch and Morris’; Hooper*; Krickhaus®; Howard®; Rupp and Lehmann’; Leh- mann’; Herroun’; Weller!®; Knorre!!; Yockey!?; Kolb and Formhals*’. 1 Pharm. J. Trans., 1879, 3rd ser., 10: 441. 2 J. Soc. Dyers Colourisis, 1896, 12: 86. 3 Z. anorg. Chem., 1900, 25: 227. 4 Institution of Mining and Metallurgy Transactions, 1908, 17: 331. 5 Eng. Mining J., 1910, 90: 357. 6 J. Am. Chem. Soc., 1908, 30: 378. 7 Apoth. Zig., 1911, 26: 203. 8 Tbid., 1912, 27: 545. ® Chem. News, 1882, 45: 101. 10 Ann., 1882, 213: 364. 1 Z. angew. Chem., 1888, 155. 12, J, Am. Chem. Soc., 1906, 28: 1435. 18 Z. anorg. Chem., 1908, 58: 189. 1920) ROARK: ARSENICAL INSECTICIDES 363 In order to test the statement of Williamson! that nitrates cause only a slight error in this method if the determination be made in the cold, amounts of sodium nitrate equal to 5 and 10 per cent, respectively, of the total sample were added to lead arsenate and the analysis carried through as directed in the 1915 report on insecticides?. The same series was also run through after evaporation to complete dryness with hydro- chloric acid. Results were as follows: TABLE 4. Effect of nitrates upon the determination of pentavalent arsenic by Naylor’s method. N/20 THIOSULPHATE USED SODIUM NITRATE PRESENT Not evaporated Evaporated per cent cc. ce. 0.00 65.30 65.30 65.30 65.30 5.00 68.60 65.20 69.00 65.30 10.00 86.20 65.20 79.40 65.30 This shows that nitrates even at ordinary temperature liberate iodin from potassium iodid in hydrochloric acid solution, but that all nitric acid is expelled by evaporating to dryness on the steam bath with an excess of hydrochloric acid. Evaporation with hydrochloric acid will also decompose lead peroxid which is sometimes present in lead arsenates prepared by roasting lead oxid and arsenic trioxid. Ferric chlorid is not removed by this treatment and will affect the results, but it is never present in more than a trace in commercial lead arsenates. In Table 5 are shown results on commercial lead arsenates by the thiosulphate method proposed in 1915 and also by the modified method in which the sample is first evaporated with hydrochloric acid. _ The directions for the modified method are as follows: TOTAL ARSENIC PENTOXID. REAGENTS. Starch solution.—Prepare as directed under Paris green®. Standard iodin solution—Prepare as directed under Paris green, but calculate in terms of As.0.°. 1 J. Soc. Dyers Colourists, 1906, 12: 86. 2 J. Assoc. Official Agr. Chemisis, 1917, 3: 157. * Assoc. Official Agr. Chemists, Methods, 1916, 63. 364 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 STANDARD THIOSULPHATE SOLUTION. PREPARATION OF SOLUTION. Prepare an approximately N/20 solution as follows: Weigh 13 grams of crystallized C. P. sodium thiosulphate, dissolve in water which has been recently boiled and cooled, filter, and make to volume in a 1 liter graduated flask, using water that has been recently boiled and cooled. To standardize this solu- tion, proceed as follows: (A) Dissolve about 0.7 gram of C. P. dilead arsenate (PbHAsOx,) in 50 cc. of con- centrated hydrochloric acid in an Erlenmeyer flask. If necessary to effect solution, heat on the steam bath, keeping the flask covered with a watch glass to prevent evapora- tion of the acid. Cool to 20-25°C., add 10 cc. of potassium iodid solution (20 grams of potassium iodid per 100 cc.) and 50 cc. (or more if necessary to produce a clear solution) of ammonium chlorid solution (25 grams of ammonium chlorid per 100 cc.), and immediately titrate the liberated iodin with the thiosulphate solution, until the solution is colorless, using starch paste as an indicator near the end point. From the weight of lead hydrogen arsenate and the number of cc. of sodium thiosulphate solution used, calculate the value of the latter in terms of As,O;. (As.O; in P»HAsO,=33.11 per cent.) (B) Titrate 50 cc. of the standard iodin solution, to which has been added 50 cc. of concentrated hydrochloric acid and 10 cc. of the 20% potassium iodid solution, with the thiosulphate solution, to a colorless solution, using starch paste as an indicator near the end point, and from the ratio of the two solutions, and the value of the iodin solution in terms of AsO; calculate the value of the thiosulphate solution in terms of AsO. The values obtained by these two methods of standardization should check very closely. The value obtained by procedure (A) is to be preferred. Pure dilead arsenate may be prepared by pouring a solution of lead nitrate into a solution of potassium dihydrogen arsenate (KH2AsO,), which should be in excess. The precipitate should be collected by filtration, dissolved in the smallest possible quantity of boiling nitric acid (1 to 4), and this solution then poured into a large quantity of distilled water. The precipitate which results should be collected and dried at 110°C. DETERMINATION. Weigh an amount of the powdered sample equal to twice the amount of arsenic pentoxid to which 100 cc. of the thiosulphate solution are equivalent. Transfer to a 400 ce. beaker, add 25-30 cc. of concentrated hydrochloric acid and evaporate to com- plete dryness on the steam bath. Take up the residue in 50 cc. of concentrated hydro- chloric acid, warming to effect solution (beaker must be kept covered to prevent evaporation), and proceed as directed under standardization (A). The number of cc. of thiosulphate solution used in the titration, divided by 2, represents directly the per cent of arsenic pentoxid in the sample. The great value of this method lies in the saving of time effected by its use. The author has found that a number of determinations may — be carried through, from weighing out the samples to entering the results, — in an average time of 8 minutes for each determination. This is exclu- sive of the time necessary to evaporate to dryness with hydrochloric acid, but this evaporation may easily be done overnight, as it does not require any attention on the part of the analyst. 1920) MANUFACTURER BSS 33H FRR See Poooo AANNAMQ ANANM AAAS ROARK: ARSENICAL INSECTICIDES 365 TABLE 5. Results on commercial lead arsenates (dried samples). z Zz 25 2 g 35 Zeso =3 se £ = rae 75 5 Biz Oe o8 3 é 5 | g2 | gz | 22 |22e6| 23 | 28 = < ° on og BS ae as) m5 & = 2 o< o< BZ Aos on Qz g g 2 oe S23 ES |fazn2| a6 a 5 3 < 4 < E g 2 5 percent | percent | percent | percent | percent | per cent | per cenl Paste 15952 | 3.54 ss eee pane oe eres | ee Paste 23169 | 1.96 we =p © —ee af an. Ped 2 ee Paste 13890 ; 0.00 Epa f- ee com 2228 Beet eee Paste 22799 | 0.39 ar ——— — S43 2 eee’ | See Paste 22994 | 0.02 See ae Bz sees eos) | ee Paste 23133 | 0.02 29.55 | 28.00 | 1.55 27.90 | 28.03 | —0.13 Paste 23218 | 0.02 ne ee eas “ve — Rel seal pee Paste 23414 | 0.16 29.48 | 27.85 | 1.63 27.80 | 28.03 | —0.23 Powder 8032 | 0.00 ae ae a ee =e |) sey Powder | 19136 | 0.20 Seen Be 2 eee bays SPS | eee eee Powder | 23132 | 0.00 says a eat Aves een oS Powder | 23134 | 0.00 weer Lae esi enna Saeed oe Powder | 23224 | 0.37 B papal Bee se ee 2peeas| i eee Powder | 23608 | 0.06 2333 =e ces —- ee Paste 20360 | 0.06 a2 S— 2 cae See eee | ee Sees Paste 22854 | 0.00 seks ——— eS Sees ee Ty Ba Paste 23239 | 0.00 Late Bese re apt 2 S Se 2 || Ee ee Paste 23347 | 0.00 ae Bees Ss ep aet Seal) Ee SS 2 Paste 13889 | 0.00 ae —— — — ees bee Paste 19231 | 0.00 29.13 | 28.28 | 0.85 28.23 | 28.28 | —0.05 Paste 22885 | 0.18 ne BE a aa Shee oa eed | Pe ec Paste 22886 | 0.50 ae pa she eee aes eA Se ee Paste 22888 | 0.50 pe = eel Ae sy Dh See Paste 22971 | 0.61 29.34 | 28.93 | 0.41 29.53 | 29.64 | —0.11 Paste 23033 | 0.37 ee =é< ve eee a ae ae Paste 23236 | 0.07 a AA ee es tee Bs |||" Saree Paste 23346 | 0.10 30.78 | 30.03 | 0.75 30.55 | 30.15 | +0.40 Powder | 19312 | 0.00 ae sae "eZ ares pe |e sae Se Powder | 21079 | 0.02 32.60 | 31.25 | 1.35 31.42 | 31.28 | +0.14 Powder | 22887 | 0.00 tgp Le ope eee See a Powder | 23032 | 0.00 Pend. 22 per: res eS oe ae Powder | 23412 | 0.13 29.98 | 28.68 | 1.30 28.95 | 28.83 | +0.12 Paste 19264 | 0.13 28.23 | 27.88 | 0.35 27.95 | 28.03 ;} —0.08 Paste 22911 | 0.12 wees seve sae aes Sir Wy oa ee Powder | 17726 | 0.05 se ‘Eck aes ee ane oer Powder | 17989 | 0.02 Bee nae See ee hal ae ae Powder | 20479 | 0.00 32.78 | 32.23. | 10:50 32.33 | 32.23 | +0.10 Powder | 22417 | 0.15 Lee Lee a pies Seem $e Powder | 22795 | 0.17 a =a —— ey eee |) poe Powder | 22910 | 0.50 = ae Sees penal eda oh | a eae 366 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TasLe 5.—Continued. ae = c & . Sis < oh a = S Zo 32 = PE ge $ a E e era eee 5a 8z = = z ES ES 38 Sok Se = i=} = ° ° Zoo ro] z g : ding aes) 8 | SE as) Gee x Q 2) & & ra Zea = = nz eS 22 og : z z Beal ue cadh ce! Wee Mesa ls ae 5 2 z i aig fe 88 |geas| 85 28 5, A S x Zo Zo En <59038 a8 Qe S = 3 cS) oe of a2 iar «5 aS : g 5 owes; |S siiliP ibe hae k ailiicshy lene z g g ¢ oa $a EX | sae2| 26 aS = ° 4 < < < a 5 a ) per cent | percent | percent | percent | per cent | percent | per cent Q Powder | 23011 | 0.02 aes | ae = eee eee | eee Q Powder | 23415 | 0.00 aoe case Soh Bhs i K whii) eee R Paste 23183 | 0.22 32.20 | 31.70 | 0.50 32.15 | 31.96 | +0.19 R Paste 23243 | 0.27 32.58 | 32.25 | 0:33 32.20 | 32.56 | —0.36 R Paste 23367 | 0.04 sat a xa. bets ED | oes R Paste 23408 | 0.29 ale oe see Bee ee eee R Paste 23494 | 0.27 oe aaah es Spe se Bee ee Ss Powder | 17716 | 0.00 ae ae ae Eee. awe Pees | Po eee ) Powder | 21165 | 0.02 33.10 | 32.15 | 0.95 31.55 | 32.17 | —0.62 Ak Paste 23223 | 0.00 pes AL —_ — Bee |e ee T Paste 23420 | 0.00 | = a he Ae hile So} ey ay Powder | 17730 | 0.00 30.85 | 29.83 | 1.02 30.00 | 29.83 | +0.17 Aly Powder | 23159 | 0.00 3: 3 oe eee, i 8 ye Nh Ee S U Paste 11551 | .0.00 23:00) } 22510) |e 21.83 | 22.10 | —0.27 \"4 Paste 23157 | 0.03 bohe Bess 2 Jes oe PSA Cet Vi Paste 23550 | 0.05 pet aes = Oe pit |e yp 2, Vi Paste 23636 | 0.04 ene = 3 Hi § 28: aris) Ke Teal Vi Paste 23657 | 0.16 ae yah ee: 2 ae SE | eee Mf Powder | 23158 | 0.00 = aed ee eau Boe fe Sa eae WwW Paste 23519 | 0.00 lye ee oer 7 2 eee | It is seen that the difference in the amount of As.O; determined with- out and with evaporation with hydrochloric acid (the modified method) is: minimum, 0.25 per cent; maximum, 5.95 per cent; average, 1.04 per cent, being always higher in the method without evaporation, due to the presence of nitrates or peroxids. The difference in the total arsenic determined by the modified Gooch- Browning method and the sum of the separate determinations of As.O; and AsO; (the latter being determined by the modified or evaporation method), both being calculated as As»O;, is: minimum, 0.02 per cent; Maximum, 0.62 per cent; average, 0.21 per cent. In practically all cases where the difference is as much as 0.2 per cent the sum of the two forms of arsenic is greater than the total determined directly. This is due to the presence of ferric salts in the sample, causing high results for As,0;. The fact that the average discrepancy is only 0.21 per cent shows that the proposed method for As,O; is sufficiently accurate for use as a tentative method for the rapid examination of commercial samples. 368 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 SUMMARY. (1) A method is proposed for the determination of As.O; in lead arsenate which has been shown to give excellent results on known mix- tures of lead arsenate and lead arsenite. Of one hundred commercial lead arsenates tested by this method, representing the product of twenty-three leading manufacturers, thirty-eight contained no As»03; forty-seven contained less than 0.30 per cent of As.O; (approximately 1 per cent of the total arsenic present); while fifteen containedAs.O; in amounts ranging from 0.30 to 3.54 percent. If all the As.O; is combined as lead arsenite, Pb;(AsO;)2, then twenty-one of the samples contain as much as 1 per cent of this compound. (2) Methods for the determination of arsenic and antimony based on the reaction As.O;+4HI= As.O;+ 21.4+2H.0 in acid solution and titra- tion of the liberated iodin with standard thiosulphate solution have been tested, and a modification is proposed that is directly applicable to com- mercial lead arsenates with an average error of less than 0.2 per cent of AsoOs. (3) Ten commercial zinc arsenites, representing the product of three manufacturers, were tested for AssO;, which was found in every case, the amounts ranging from 0.32 to 2.05 per cent. In many cases this AsoO; is present as lead arsenate (lead being nearly always present in commercial zine arsenites), while in some instances it is really Sb.Os that is present although reported as As2Os. REPORT ON WATER. By W. W. Sxinner (Bureau of Chemistry, Washington, D. C.), Referee. During the current year there was developed and published by Samuel Palkin! of the Bureau of Chemistry, a method for the separation and determination of lithium, which appeared to be such a decided improve- ment over the troublesome and disagreeable Gooch amy] alcohol method, the official method of the association, that the referee decided to confine the cooperative work to a thorough test of the accuracy of the new method for the determination of lithium and to determine the effect of the ether-alcohol separation, if any, upon the accuracy of the method for the determination of potassium and sodium. The method depends upon the use of absolute alcohol and ether for the separation of the lithium chlorid from the mixed chlorids, and Palkin found that by — observing certain precautions, the separation was so complete that the use of a factor of correction for the solubility of sodium chlorid and potassium chlorid was entirely obviated. This is a decided improve- 1 J. Am. Chem. Soc., 1916, 38: 2326. 1920) SKINNER: REPORT ON WATER 369 ment over the amy! alcohol method, in which a factor must be used, and which is likely to be a cause of controversy, especially when, as is not unusual in the determination of lithium in mineral waters, the lithium found is of the same order of magnitude as the factor of correction. The method is as follows: The total alkali chlorids are dissolved in a minimum amount of cold water in a tall 200 ec. beaker. About 1.5 cc. will be more than sufficient for 0.5 gram of the salts. One drop of concentrated hydrochloric acid is added and gradually 20 cc. of absolute alcohol, the alcohol being dropped into the center of the beaker (not on the sides) while rotating. The sodium and potassium chlorids should be precipitated in a per- fectly uniform granular condition. In a similar manner, while rotating the beaker, 60 ec. of ether (sp. gr. at 25°C., 0.716—0.717) are added and the mixture is allowed to stand about 5 minutes, or until the precipitate is well agglomerated and the supernatant liquid almost clear. The beaker is rotated occasionally. The mixture is then filtered through a weighed Gooch crucible into an Erlenmeyer flask, using a bell-jar arrangement. The beaker is thoroughly washed with a mixture of 1 part alcohol and 4-5 parts ether. A rubber-tipped rod is necessary for this pur- pose. The precipitate in the Gooch crucible is also well washed and the crucible set aside. The funnel is well washed to remove any lithium therefrom into the flask containing the filtrate. The filtrate is evaporated to dryness on the steam bath (using a blast). The residue is taken up with 10 cc. of absolute alcohol, warming if necessary, so that practically everything passes into solution. If a slight film remains on the bottom of the flask and sides, it is removed by rubbing with a rubber-tipped glass rod. While rotating the flask, 50 cc. of ether (sp. gr. at 25°C., 0.716-0.717) are added. One drop of con- centrated hydrochloric acid is added, the flask rotated and allowed to stand for 30 minutes. It is well to rotate the flask at frequent intervals. When the fine precipi- tate has agglomerated (only a very small amount is usually precipitated), it is filtered through the crucible used in the first precipitation into a tall beaker, a bell-jar arrange- ment being employed. The residue is washed with ether-alcohol mixture, using the Same precautions as outlined in the first precipitation. After drying in an oven, the crucible is gently ignited, cooled and weighed. The ether-alcohol solution of lithium is evaporated on the steam bath. The residue is taken up in a little water and a slight excess of sulphuric acid added. The solution is then carefully transferred to a weighed porcelain or platinum dish, the solution is evaporated as far as possible on the steam bath, and the residue then gently ignited over a flame. By placing the dish on a triangle over an asbestos gauze and using a low flame, the solution can be evaporated without spattering. The residue is then carefully ignited over a full flame. When charring has occurred, it is well to repeat the ignition with sulphuric acid. Calculate to lithium, using the factor 0.12625. Remove the chlorids of sodium and potassium from the Gooch crucible with 25-50 ee. of hot water, collecting the filtrate in a porcelain dish by means of the bell-jar ‘arrangement. Add sufficient platinic chlorid solution (containing the equivalent of 1 gram of metallic platinum, i. e., 2.1 grams H,PtCl, in every 10 cc.) to convert sodium and potassium to their respective double chlorids and evaporate to dryness. Treat the residue with 80% alcohol, filter, and wash until the excess of platinic chlorid and sodium platinic chlorid has been removed. Dry the filter and precipitate, dissolve the residue in hot water, and transfer to a weighed platinum dish. Evaporate on the steam bath, dry for 30 minutes in the oven at 100°C. and weigh as potassium platinic chlorid; 370 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [| Vol. III, No. 3 calculate to potassium chlorid, using the factor 0.30573; and to potassium, using the factor 0.16085. Find the weight of sodium chlorid by subtracting the weight of potassium chlorid from the total weight of the chlorids obtained above. Calculate to sodium, using the factor 0.39343. Report as milligrams of sodium, milligrams of potassium, and milli- grams of lithium per 50 cc. of solution. Two samples were prepared containing known amounts of lithium, potassium and sodium chlorids. Sample 1 contained 1.3 mg. of lithium per aliquot taken for analysis and was assumed to represent an amount which might be expected in a lightly mineralized sample of water. Sample 2 contained 25.7 mg. of lithium per aliquot taken and was assumed to represent about the maximum amount which might be ex- — pected in a sample of water taken for analysis. The samples were sent out to nine analysts who had asked to partici- pate in the cooperative work. Reports have been received from six. Cooperative work on water. MiIxuiGraMs IN 50 cc. ANALYST LITHIUM POTASSIUM SODIUM Sample | Sample || Sample | Sample |! Sample | Sample 1 2 1 2 1 2 W. D. Richardson, Swift & WS 25.8 26.4 32.4 77.8 93.6 Company, Chicago, Ill. 1.8 25.8 26.4 32.3 77.4 93.2 1.4 25.9 26.5 32.1 76.9 92.8 Samuel Palkin, Bureau of Chem- 1.6 25.4 26.3 31.5 78.1 94.4 istry, Washington, D. C 1:5 Dot 26.4 og 77.9 ae me 25.7 ae 31.8 —— 94.0 A. N. Bennett, State Water Sur- 1.5 Dot 26.6 31.9 78.1 94.0 vey, Urbana, Ill. 1.5 20.2 26.6 32.0 78.4 93.8 1.5 25.3 Se 32.0 ome 94.0 J. W. Sale, Bureau of Chemistry, 1.4 25.7 27.6 33.5 76.5 92.1 Washington, D. C 1.4 25.6 27.8 33.4 76.4 93.0 1.3 25.5 27.6 33.6 cae 92.5 R. H. Kellner, Bureau of Chem- 1.3 25.6 26.7 32.4 77.2 92.5 istry, Washington, D. C 1,2 25.5 26.9 32.1 77.2 92.8 1.3 25.6 26.8 31.9 77.3 92.6 W. F. Baughman, Bureau of 1.4 25.5 27.0 31.9 77.3 93.7 Chemistry, Washington, D. C. 13 25.3 26.7 32.1 77.5 93.6 fo Sa seas ES EES UA __ AVGTE RO) j= n> eens mes 1.4 25.5 26.8 32.3 77.4 93.3 Miphest. a. eee eee sig 1.8 25.9 27.8 33.6 78.4 94.4 Lowest. «SOU ee Fe 1.2 25:1 26.3 31.5 76.4 92.1 Theory __ - _- 28 32h soe os be 1.3 25.7 26.2 31.4 77.9 93.4 From the tabulated statement it will be noted that the results for lithium by the Palkin method are all that could be desired, but that the | 1920] VEITCH: LIME REQUIREMENT OF SOILS orl potassium results are consistently high, the average of the work reported by the six analysts being approximately 2.5 per cent higher than the theoretical amount present. It should be noted further that while the variation in potassium amounted to 2.5 per cent of the theoretical amount present, the actual error in weight amounted to only 0.8 mg., an error which for most analytical purposes may be regarded as negligible. The results for lithium are so entirely satisfactory that the referee feels warranted in recommending the method for adoption by the association as an optional official method. RECOMMENDATION. The referee recommends the adoption of the method for the determina- tion of lithium, potassium and sodium (per method in the body of the report) as an official method. This is the first recommendation of a referee on this method, submitted as provided in By-law No. 6 of the association. The method has not been published heretofore in the proceedings. REPORT ON THE LIME REQUIREMENT OF SOILS. By. F. P. Verrcu (Bureau of Chemistry, Washington, D. C.), Referee. It is regretted that only a report of progress can be made at this time. Known samples have been secured from the Rhode Island, Pennsylvania, and Maryland station plats, as well as from other soils the general history of which is known for twenty or more years. Samples have also been obtained upon which other investigators have worked, using other methods for determining lime requirements. Large quantities of some of the samples are on hand, and it is proposed to retain these for testing by the methods which may be developed in the future. The referee will be pleased to supply portions from these samples to those who are developing new methods. The referee has proposed previously that the reaction to phenol- phthalein of distilled water which has been in contact with the soil, with frequent shaking, from 16 to 20 hours, be used as the basis for distinguishing acid and basic soils. This standard is based on his experience and observations that nearly all soils which, under general farming conditions, bore a thick, strong, dark-colored stand of red clover, gave a basic reaction by this procedure. Later experience indicates that if the water has a basic reaction to delicate red litmus paper, it may be considered basic even though it is not basic to phenolphthalein. Holman! observed that when the soil extract obtained with the lime 1 Science, 1916, new ser. 44: 311. 372 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 water method is colored with dissolved organic matter, the color pro- duced on adding a phenolphthalein solution to the boiling extract fre- quently fades out if the boiling is continued, but that the extract is positively basic to red litmus paper. Several investigators, including the referee, have suggested that the lime water method originally proposed gives high results on soils rich in organic matter. The above mentioned fact is one cause for this. Another is the fact, also observed by Holman, that the calcium carbonate dis- solved from the soil by the distilled water in contact with it, frequently does not diffuse into the upper layers of the supernatant liquid even on standing overnight. This is especially true of those soils which are but weakly basic in reaction. These two difficulties are eliminated by the modifications proposed in 19161. During the past year the referee’s work has been limited to a further study of the lime water method. A number of indicators have been studied to determine which gives the most marked and definite end point with soil extracts. Among those examined were alizarin, cyanin, rosolic acid, tetrabromphenolsulphonephthalein, dinitrobenzoylene urea, paranitrophenol, phenolsulphonephthalein, and thymolsulphonephthalein. None of the reagents has proved so satisfactory in the referee’s hands as carefully washed red litmus paper, which is more sensitive and more nearly indicates exact neutrality than any of the others and is also much more convenient to use. The lime water method as now operated by the referee is as follows: LIME WATER METHOD FOR DETERMINING THE LIME REQUIREMENTS OF SOILS. APPARATUS AND REAGENTS. The suitability of the glassware, water, indicators, and filter paper must be definitely determined by careful trial. For this purpose place a filter paper in a well-washed funnel and wash it thoroughly 3-5 times with the distilled water. Run through the paper 100 cc. of the distilled water, receiving it in a well-washed beaker which is to be tested. Add to the filtrate 1 or 2 drops of phenolphthalein solution, a piece of the red litmus paper and boil to a volume of 5 cc. There should be no alkaline reaction. Add 1-2 drops of the standard lime water. The reaction to both the indicators should be alkaline. DETERMINATION OF SOIL REACTION, Place 10-12 grams of soil in a 100 cc. non-soluble glass flask, add 80-100 ce. of the tested neutral distilled water, stopper, shake thoroughly, and allow to stand 16-18 hours, shake, filter through a well-washed neutral 12.5 em. filter (No. 588, C. S. & 5. , folded filter paper is good), rejecting the first 10 cc. of the filtrate and returning the filtrate until practically clear. Place 50-60 ce. of the filtrate in a clean 100 cc. beaker of non-soluble glass and boil to a volume of about 10ce. Remove from the gauze, place 1 Setence, 1916, new ser. 44: 311. NN er aaa ee ee 1920] VEITCH: LIME REQUIREMENT OF SOILS 373 the beaker on a piece of white paper and add 2 drops of the phenolphthalein solution and carefully observe the color. If a pink color, no matter how faint, develops, the soil is basic. Ifa pink color does not develop, add a small piece of delicate red litmus paper and allow to stand for 10 minutes. If the litmus paper turns blue, the soil is basic. If it remains red or reddish purple, the soil is acid. DETERMINATION OF LIME REQUIREMENT. To severai 11.8—12.0 gram portions of acid soil (as indicated by the concentration of the standard lime water solution which varies from 1.18—1.20 grams of calcium oxid per liter) in 100 cc. flasks of non-soluble glass, add 10-15 cc. of distilled water and different quantities of lime water, arbitrarily selected (according to the nature of the soil) and differing from each other by 5 or 10 ce. Close with clean cork stoppers and allow to stand at room temperature with frequent shaking for 2 hours. Dilute to 75 or 80 cc. with distilled water, and allow to stand with frequent shaking for 2 hours. Filter and test the reaction as described under ‘Determination of Soil Reaction”. If any extract is alkaline to either or both indicators, that portion of soil has been rendered basic by the added lime water. If no portion gives a basic reaction, these preliminary tests must be repeated with larger quantities of lime water. The largest quantity still giving an acid reaction and the smallest quantity giving a basic reaction establish limits within which lies the lime requirement of the soil. With this information as a guide, take fresh portions of the soil and repeat the tests with quantities of lime water differing from each other by 1 or 2 cc. and lying within the limits established by the previous tests. The smallest amount of lime water which gives the characteristic pink or blue color is taken as the lime requirement of the soil. Each cc. of standard lime water is equivalent to a lime (CaO) requirement of 0.01 per cent or of 100 parts of lime (CaO) per 1,000,000 parts of soil. This procedure has resulted from the trial on numerous samples of several modifications, including treating the soil with the lime water at 60°C. for 2 hours, diluting and allowing to stand from 14 to 16 hours, moistening basic soils with distilled water only, and drying on the steam bath; heating for 2 hours at 60°C., diluting and allowing to stand from 14 to 16 hours. These procedures all gave higher results on soils rich in organic matter. Some basic soils even appeared to be acid and to have a marked lime requirement. Clays and loams containing no organic matter gave closely agreeing results by all procedures when the final time of stand- ing in contact with distilled water was the same in all cases. It must be remembered, however, that the longer the soil stands in contact with distilled water the lower its apparent lime requirement. Many soils which are basic to the procedure above described for determining the reaction of soils are decidedly acid and have a high lime requirement when tested by the hydrogen electrode procedure sug- gested by Sharp and Hoagland', or by the freezing point procedure proposed by Bouyoucos?. It is desired to point out that this method does not indicate the quan- 1 J. Agr. Research, 1916, 7: 123. _? Mich. Agr. Expt. Sta. Tech. Bull. 27. 374 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 tity of lime necessary to maintain the basic reaction of a soil throughout a growing season or under field or cropping conditions. No method for determining lime requirement can do this. The method is designed to determine the lime requirement of the sample at the time it is examined. Assuming that the sample is repre- sentative, the additional quantity of lime which will be required to maintain a basic reaction in any soil under any given conditions of tex- ture, crop, organic matter, moisture, bacterial activity, etc., can only be estimated. It can not be definitely or even approximately determined. L. P. Howard (Agricultural Experiment Station, Kingston, R. I.), submitted a paper on ““The Relation of the Lime Requirements of Soils to Their Retention of Ammonia”! ?. DRUG SECTION. No report on medicinal plants and drugs was made by the referee. REPORT ON SYNTHETIC PRODUCTS. By W. O. Emery (Bureau of Chemistry, Washington, D. C.), Associate Referee. An adapted method for estimating hexamethylenetetramin in phar- maceutical practice, namely, tablets, was sent to the collaborators. The method, as developed on numerous controls, took the following form: REAGENTS. A. Modified Nessler’s reagent.—(a) Solution of 10 grams of mercuric chlorid, 30 grams of potassium iodid and 5 grams of acacia in 200 cc. of water, filtered through a pledget of cotton; and (b), solution of 15 grams of sodium hydroxid in 100 ce. of water. B. N/10 iodin solution. C. N/20 thiosulphate solution. PRELIMINARY TREATMENT. Ascertain the weight of 20 or more tablets (Sample 19), triturate in a mortar to a fine powder and keep in a small capsule tightly closed with a cork or glass stopper. Weigh out 0.5 gram (1 gram in the case of Sample 20, which consists of about equal — parts of hexamethylenetetramin and talc) of the powdered product on a metal scoop — or watch glass, transfer with sufficient water to a round-bottomed flask, add additional water to a total volume of 100 ce. and finally 25 cc. of 10% hydrochloric acid. Connect with a reflux condenser (preferably of the worm type) and boil gently for 15 minutes. — Cool, wash out the condenser tube with a little water and transfer the contents of the flask quantitatively to a graduated 250 ce. flask, finally diluting to the mark with water. 1 Soil Science, 1918, 6: 405. ? Presented by B. L. Hartwell. 1920] EMERY: REPORT ON SYNTHETIC PRODUCTS 375 METHOD. With a pipette withdraw 10 cc. (containing, in the case of a pure product, the ele- ments of 0.92 gram of hexamethylenetetramin) of the solution so prepared to a 200 cc. Erlenmeyer flask containing a mixture (chilled in ice-water if available) of 20 cc. of reagent A (a), and 10 cc. of A (b), wash down the neck of the container with a jet of water from the wash bottle and allow to stand for at least 1 minute. Add 10 cc. of 40% acetic acid in such a manner that the inside of the neck is completely washed by the reagent, mix quickly and thoroughly by rotating and tilting the flask, and immedi- ately run in from a burette 20 cc. of solution B, then titrate with C, adding 5-10 drops of starch solution toward the end of the operation, to the disappearance of the blue coloration. The final color of the solution is a pale straw-green. If preferred, the end point may be determined by the reappearance of a faint blue coloration, induced by the addition of a drop of iodin. COMMENTS AND SUGGESTIONS. The foregoing is essentially a reversal of the procedure first employed by Rupp! in the evaluation of mercuric chlorid tablets with formalde- hyde and subsequently made use of by Stiiwe? in the estimation of for- maldehyde, formalin and hexamethylenetetramin. Unfortunately, none of the earlier operators apparently felt the neces- sity of describing in detail the several steps followed by them, hence the writer was compelled to determine for himself the salient factors bearing particularly on the quantitative side. Briefly, the method involves four principal operations, namely: (1) Hydrolysis of the hexamethylenetetramin to formaldehyde and ammonia; (2) interaction of formaldehyde with potassium mercuric iodid of reagent A (a) (1); (3) solution of the mercury resulting there- from; and, (4) titration of the unexpended iodin. From the data thus gained, the quantity of hexamethylenetetramin is readily calculated. The reaction taking place between formaldehyde and potassium mer- curic iodid in the presence of caustic alkali is given in the equation: CH,O+K.HgI,+3KOH = Hg+HCO.K +4K1+42H,0. In conducting the method proper, unusual care is necessary in adding and mixing the several reagents with the preceding menstruum so that a uniform solution will result. This is effected by judicious rotation and tilting of the flask, and at certain points also by washing down the neck of the container with a fine jet of water. The addition of iodin should follow acidification with the greatest possible dispatch, because long standing of the mixture in the presence of free acetic acid invariably leads to low values for hexamethylenetetramin, due apparently to partial resolution of the colloidal mercury first precipitated. The primary chilling of Nessler’s reagent is advocated in order to minimize to the utmost any tendency toward secondary reactions, and also to avoid the 1 Arch. Pharm., 1906, 244: 540; 1914, 252: 430. * Tbid., 1905, 243: 300; Pharm. Ztg., 1914, 59: 215. 376 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 possible loss of iodin through undue increase in temperature on the addition of acetic acid. Since the standard iodin (reagent B) has twice the strength of the thio- sulphate (reagent C), and 1 cc. of N/10iodin is equivalent to 0.001167 gram of hexamethylenetetramin (O=16), the quantity of this product in the aliquot under examination may be readily calculated from the expression: HN 0.001167 in which H=number of cubic centimeters of reagent C, equivalent to 20 cc. of reagent B; I=number of cubic centimeters of reagent C required to offset the unexpended iodin; and N=the normality of reagent B. Before formulation and submission of the foregoing method for col- laborative purposes, many experiments were made to the end that any factors, such as the influence of time, temperature, concentration, vehicles and diluents employed in the manufacture of tablets, calcu- lated to affect unfavorably the quantitative results, might be deter- mined and subsequently eliminated, or, as far as possible, counteracted. Thus, it was found that precipitation of colloidal mercury is practi- cally instantaneous and hence complete after the lapse of one minute from the time the mixture has attained homogeneity. However, since secondary reactions at this point are not to be feared, there can be no objection to allowing the mixture to stand for a longer period, if desired, before the addition of acetic acid. In order to ascertain to what extent, if any, the final result might be affected by varying the time during which the precipitated mercury is subjected to the solvent action of acetic acid, a series of controls was carried out involving 15 seconds, 1, 2, 3, 4, 5, 10, 20 and 30 minute intervals, on the lapse of which the mixtures were treated immediately with iodin and then titrated as prescribed in the method. The recoveries of hexamethylenetetramin in these tests were 100.1, 99.5, 98.9, 98.5, 97.9, 97.7, 97.3, 96.7 and 96.3 per cent, respectively, thereby showing that protracted contact of the colloidal metal with the acid invariably leads to low values. That the presence of vehicles or diluents like starch, lactose and acacia has no appreciable effect on the final outcome is clearly shown in recoveries of 99.8, 99.9 and 100.2 per cent of hexamethylenetetramin, respectively, obtained with controls involving the substances in ques- tion; hence their elimination from tablets prior to analysis is entirely unnecessary. Of the dozen or more chemists who indicated a willingness to col- laborate, eleven submitted their findings in time for incorporation in 1920| EMERY: REPORT ON SYNTHETIC PRODUCTS Sid the present report. The data so presented have been arranged in the following tabulated form: Cooperative results on synthetic products. CONTROL ANALYST METHYLENE-| NO.19 | NO. 20° TETRAMIN per cent per cent per cent Ant. rene Bureau of Chemistry, Washington, [DL Ope Se es eee ernie ts eet ror 99.8 92.8 50.2 V. B. Bonney, U. S. Food and Drug Inspection Sta- tion, U. S. Appraiser’s Stores, San Francisco, Cal. ___ apes 90.5 50.0 ‘L. A. Brown, Agricultural Experiment Station, Lex- “ag hetty [Rep ee pa ie 99.9 91.6 49.3 J. F. Darling, U. S. Food and Drug Inspection Sta- tion, U. S. Appraiser’s Stores, New York, N. Y.___- ea 88.4 50.0 W. O. Emery, Bureau of Chemistry, Washington, D. C. 99.8 90.7 50.8 C. K. Glycart, U. S. Food and Drug Inspection Sta- tion, Transportation Building, Chicago, [ll._______~- seea 90.4 50.5 W. S. Hubbard, Bureau of Chemistry, Washington, nee rer oe ee ES ee eee ae 90.6 49.9 H. B. Mead, U.S. Food and Drug Inspection Station, U. S. Appraiser’s Stores, Philadelphia, Pa._______~ 98.3 90.7 63 yf C. B. Morison, Agricultural Experiment Station, ewshHavensGonns 22 fF < Sei if. yA BEAL fee 90.6 50.1 C. E. Parker, U.S. Food and Drug Inspection Station, Tabor Opera House Building, Denver, Colo. _-__--- 99.6 91.9 50.6 W. R. Rich, Agricultural Experiment Station, Orono, WIS. = eee ees 93.3 50.4 C. D. Wright, Bureau of Chemistry, Washington, D. C. 99.9 91.9 50.6 As above indicated, controls were carried out by some of the workers on various standard brands of the pure product. Number 19 consisted of a certain brand of commercial tablets containing 10 per cent more or less of a vehicle or diluent. Number 20, on the other hand, was a lab- oratory product obtained by triturating talc and hexamethylenetetramin in a mortar in about equal parts, and presumably, therefore, a more uniform preparation than Number 19. The foregoing percentages are for the most part averages drawn from two or more determinations carried out, if not strictly, in general accordance with the method. Some of the collaborators, however, found time to make additional tests 378 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 with a view to betterment in detail, or to ascertain to what extent, if any, changes in the matter of time, temperature, size of aliquot, etc., might affect the final outcome. As the result of a very interesting series of experiments involving modifications in time and temperature, Mr. Brown concludes that the reaction is not complete under 4 minutes, and that a temperature below 20°C. has a tendency to give low results even when the time is extended to 4 minutes. Accordingly, he believes that the method should read: ‘Allow the reaction to proceed for at least 4 minutes at a tempera- ture of 20-25°C.” As a matter of curiosity, Mr. Darling fumed a portion of each sample with sulphuric acid, later estimating the ammonia resulting therefrom. The values obtained for hexamethylenetetramin agreed quite well with those reported above. On account of the relatively small amount of substance represented in the aliquot used, Mr. Mead tried tripling the quantity of hydrolyzed product. Practically the same recoveries were obtained as in the regular way, thus apparently indicating that no greater accuracy is secured by increasing the amount of substance prescribed in the method. In connection with Mr. Morison’s report, the following note was added: Five-tenths gram of the pure salt was hydrolyzed by bciling with dilute acid under a reflux condenser, essentially as proposed in the cooperative method. The acid solu- tion was then cooled and brought to a volume of 250 cc. An aliquot was transferred to an Erlenmeyer flask and mixed with 25 cc. of approximately normal sodium hydroxid solution, followed by the addition of 25 cc. of N/10 iodin. The resulting mixture was allowed to stand for about 10 minutes to insure completion of the reaction, and then acidified with 30 cc. of approximately normal sulphuric acid. The liberated iodin was thereupon titrated with N/20 thiosulphate. 1 cc. of N/10 iodin=0.001167 gram of hexamethylenetetramin. This is an application of the G. Romijn method for formaldehyde. Recoveries by this method were as follows: 20 mg. taken; recovered in four determinations, 20.1, 20.5, 20.2 and 20.3 mg. Im a single determination on No. 20, 49.8 per cent recovery was effected. It is believed that further study should be made of the method as well as of the procedure outlined by Mr. Morison. No report on medicated soft drinks was presented by the associate referee. No report on balsams and gum resins was presented by the associate referee. 1920) FULLER: REPORT ON ALKALOIDS 379 REPORT ON ALKALOIDS. By H. C. Futter (Institute of Industrial Research, Washington, D. C.), Associate Referee. The work included the study of methods for determining atropin in tablets and strychnin in elixirs. The method used for strychnin last year was further studied and a recommendation for its provisional adop- tion is included in this report. ATROPIN. DETERMINATION OF ATROPIN IN TABLETS. Weigh 25 tablets and introduce directly into a small separator. Moisten with 5 cc. of water. Add 1 cc. of stronger ammonia water. Agitate with 25 cc. of chloroform and allow to stand until separation is complete. Draw off the chloroform into a second separator and repeat the agitation twice more with 25 cc. portions of the solvent. After combining all of the fractions, wash the combined chloroform solutions by agitation with 10 cc. of water and allow to stand 15 minutes. Introduce a pledget of absorbent cotton into the stem of the separator and run off the chloroform into the tared dish, but do not allow the wash water to enter the orifice of the stop-cock. Add 10 cc. of chloroform, and when the water has entirely risen to the surface run off the chloroform into the tared beaker. Wash off the outer surface of the stem of the separator with a little chloroform and then evaporate over a water bath, using a fan or blower and remoy- ing from the bath as the last portions evaporate to avoid decrepitation. Check the weight of the atropin by dissolving the residue in neutral alcohol, adding an excess of N/10 sulphuric acid and titrating back with N/50 potassium hydroxid. Calculate to atropin sulphate (1 ec. of N/50 H2,SO,=0.005741 gram of atropin). Factor for atropin to atropin sulphate, 1.1695. Cooperative results on atropin sulphate. ATROPIN SULPHATE ANALYST . Percent | Grain per 2.29 0.01 J. R. Eoff, Bureau of Internal Revenue, Washington, D. C.__- 0.0095* A. W. Hanson, U.S. Food and Drug Inspection Station, Trans- 2.00 0.0087 portation Building, Chicago, Ill. 1.23 0.0053" H.C. Fuller, Institute of Industrial Research, Washington, D.C. 1.99 0.008 ® By titration. STRYCHNIN. DETERMINATION OF STRYCHNIN IN ELIXIR OF IRON AND STRYCHNIN. _ Tare a 50 cc. volumetric flask, fill to mark with sample and weigh. Pour into an evaporating dish, wash the flask with water and evaporate the alcohol. Transfer to an 8-ounce Squibb separator. Add an excess of ammonia. Agitate with 25 cc. of chloroform and allow to stand until separation is complete. Draw off the chloroform 380 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 into a second separator and repeat the agitation twice more with 25 cc. portions of the solvent. After combining all of the fractions, agitate with 3 portions of 10 ce. each of N/1 sulphuric acid, collecting the acid solutions together in a fresh separator. Discard the chloroform. Treat the acid solution with an excess of ammonia, agitate with 15 cc. of chloroform and allow to stand until separation is complete. Draw off the chloroform into a second separator and repeat the agitation twice more with 15 cc. portions of chloroform. After combining all of the chloroform fractions, wash by agitation with 10 cc. of water and allow to stand 15 minutes. Introduce a pledget of absorbent cotton into the stem of the separator and run off the chloroform into a tared dish, but do not allow the wash water to enter the orifice of the stop-cock. Add 10 cc. of chloroform, and when the water has entirely risen to the surface, run off the chloroform into the tared beaker. Wash off the outer surface of the stem of the separator with a little chloroform and then evaporate over a steam water bath, using a fan or blower and removing from the bath as the last portions evaporate to avoid decrepitation. Dry at 100°C. to a constant weight and weigh as strychnin. Strychnin to strychnin sulphate, 1.2814, according to U. S. P. Cooperative results on elixir of iron and strychnin. STRYCHNIN SULPHATE ANALYST G ¥ Per cent par Parke} Davisi@ Co: Detroit, Mich?sessee= ees =e 0.037 0.187 0.035% 0.179* E. K. Nelson, Bureau of Chemistry, Washington, D. C. __-_-_- 0.033 0.166 0.034 0.17 J. B. Luther, U. S. Food and Drug Inspection Station, U.S. | 0.0342 0.17 Appraiser’s Stores, New York, N. Y. 0.0333 0.166 0.0318" 0.1568 0.031 =e W. R. Rippetoe, Schieffelin & Co., New York, N. Y.-------- 0.039 0.22 J. P. Street, Agricultural Experiment Station, New Haven, | 0.0375 0.19 Conn. 0.038" 0.194 HeiGs Moller ae ee a eee eee eee 0.040 0.20 0.036 0.18 ® By titration. RECOMMENDATIONS. It is recommended— (1) That the methods for the determination of strychnin in tablet triturates be made provisional. (2) That the method for the determination of strychnin in liquids where it occurs as the only alkaloid be made provisional. (3) That a further study be made of the method for determining atropin in tablets. (4) That the work on alkaloids be extended to a study of methods of determination of strychnin and quinin in admixture. 1920] VIEHOEVER: MEDICINAL PLANTS 381 REPORT ON MEDICINAL PLANTS. By Arno VIcEHOEVER (Bureau of Chemistry, Washington, D. C.), Associate Referee. The report is divided into three parts: I. A consideration of a method for the determination of volatile mustard oil found in true mustard and in mustard substitutes. II. A brief discussion of the determination of ethereal oil in drugs and spices. III. The adulteration of crude drugs and spices. PART I. While the new Pharmacopeeia gives a method for the determination of allylisothiocyanate in volatile mustard oil, no method has been given for the liberation of the oil from the mustard seed. The method adopted by the Pharmacopeeia for the determination of allylisothiocyanate con- tent is essentially that of the German, Swedish, French, and Japanese Pharmacopeeias. It is in principle the method of E. Dietrich', as modified by Gadamer?. Comparative investigations on the different methods undertaken by different workers? showed the advantage and disadvantage of one or the other methods sufficiently, so that no repetition of that work appeared to be warranted. In view, however, of the existing uncertainty as to certain factors which might influence the result, especially conditions of maceration, the following experiments were undertaken with Chinese colza (Brassica campestris chinoleifera Viehoever) : (1) Maceration at different temperatures. (2) Maceration at different intervals, for different periods. (3) Maceration with and without shaking the maceration mixture. (4) Maceration with and without addition of alcohol to the macera- tion liquid before the maceration. (5) Maceration with Sinapis alba (white mustard). All data indicate that lower results are obtained with prolonged maceration, and the results furthermore indicate that temperature higher than room temperature tends to hasten the liberation of mustard oil. Shaking during the maceration apparently has no decided influence on the result. The addition of alcohol to the maceration liquid before 1 Helfenberger Annalen, 1886, 1: 59. 2 Arch. Pharm., 1899, 237: 110. 3 Wehrmann, Wegener, Braunwarth and Meyer. Arch. Pharm., 1915, 253: 308. 382 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 maceration has given higher results. The question as to whether or not this higher result is due to secondary reaction, possibly the development of allylthiourethan, is under investigation. The following table gives the results of the collaborators: Influence of time and temperature of maceration upon yield of volatile mustard oil. ANALYST V.K. Chesnut, Bureau of Chemistry, Washington, DNG: . F. Darling, U.S. Food and Drug In- spection Sta- tion, U.S Appraiser’s Stores, New York, N. Y. P. L. Gowen, Bureau of Chemistry, Washington, es E. H. Grant, Bureau of Chemistry, Washington, ~u. E. K. Nelson, Bureau of Chemistry, Washington, Cc. . R. Rippetoe, TEMPERATURE TEMPERATURE TEMPERATURE (Room) 22-27°C. 37°C. 50°C. TIME TIME TIME 2 4 6 1 2 4 6 2 4 6 HOURS | HOURS | HOURS || HOUR | HOURS | HOURS | HOURS || HOURS | HOURS | HOURS OM O42 eae 0.47 | 0.45 | 0.48 } ---- 0:50) 0415 ieee L531 01421) ae res ae O43) | eee 0:48") 0:41" || 0.45 | 0.42 | 0.38 _... | 0.44 | 0.41 | 0.33 || 0.48 | 0.40 | 0.33 paisa | peter | ie ae OSLsINOG2 5 222s | ase aon | Soe eee ok eee tk. Seat OUR SERRE |] Le Seen) SS ee | eee eee 0.46 | 0.37 | 0.28 || __-_ | 0.45 | 0.32 | 0.29 || 0.46 | 0.37 | 0.28 0.46 | 0.37 | 0.26 ____ | 0.45 | 0.32 | 0.27 || 0.44 | 0.34 | 0.24 0:43] (0:63 | ‘O'49 || _-__ || (0.47 || 0:61 || O67 ||| {2 Seateeeess eee 0.48 | 0.55 | 0.63 EE OAT || OGL") "0:67, --2 Seer ee SEE O DS ylineces BP eat eee vO! Do meee --==\ |) see | eee 0.31 | 0.21 z ee a ROSS |) 10:22); ae se ae) ee 0.30 | 0.22] _-_. = Osa) O22) "eae ees ee EAS AOS: 02 eS SER Meee to ae LJ. |) Se) eee 0.40 | 0.33 | 0.25 __ | 0.41 | 0.25 | 0.36 |] 0.58 | 0.57 | 0.56 0.32 | 0.23 _--- | 0.40 | 0.29 | 0.21 0.57 | 0.53 | 0.53 and N.Smith,} 0.41 Schieffelin & Co., New York, N. Y. ® Twenty cc. of alcohol were added before maceration. The addition of Sinapis alba has been tried out, but had no effect on the result. The following table shows the effect of maceration for various periods of time with and without white mustard: 1920] VIEHOEVER: MEDICINAL PLANTS 383 Effect of white mustard, added to maceration mixture, upon yield of volatile oil. (Analysts, A. Viehoever and C. O. Ewing.) PERIOD OF MACERATION® MATERIAL AMOUNT 1 HOUR 2 HOURS 3 HOURS 4 HOURS grams per cent per cent per cent per cent White mustard alone_____________ 10 223 0.06 0.06 0.06 Chinese colza alone_____-___----_- 10 0.45 0.42 0.40 pete. White mustard and Chinese colza__ ; 5 a 0.51 0.51 0.51 of eac * Temperature of maceration, 37°C. The addition of tartaric acid had no effect, which is confirmed in the work mentioned in the above article concerning comparative studies. The powdered seed was usually put through a 1 mm. sieve and then no olive oil was necessary to prevent frothing. Finer ground flour, however, tended to froth. The best conditions for maceration appeared to be 2 hours at 37°C. Whether or not alcohol should be added to the maceration mixture before maceration has not been definitely decided. Other modifications are under consideration, e. g., the best conditions to receive the volatile oil, whether ammonia alone or mixed with alcohol or silver nitrate solution; furthermore, whether or not the distillation mixture after addition of silver nitrate solution should best be heated directly for 1 hour on the steam bath, or after standing 12 to 24 hours. Of special interest is the fact that Leach, in his book on Food Analysis, gives a factor which is about three or four times too high. The original article of Roeser has undoubtedly been misinterpreted. This method gives, as recent investigators have shown, results which agree very well with those obtatined with the Dietrich-Gadamer method. High results obtained with the Roeser method may thus often be explained, and it is suggested that statements as to abnormally high amounts found in mustards and mustard substitutes be regarded with suspicion, especially if the Roeser method has been used or if the method used has not been stated. PART II. A general method for the determination of volatile oil is very much needed. No standard has yet been adopted as to the amount of volatile oil in drugs and spices, and this naturally would be most desirable in the proper judgment of the strength and quality of such goods. A con- sideration of the method described in Bureau of Chemistry, Bulletin 107 (Revised), page 163, is of interest. While the results obtained with the method may be correct so far as the non-volatile ether extract is concerned, it is feared that during the evaporation of the ether, and 384 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 later on, especially during the drying for 18 hours in the desiccator, appre- ciable amounts of volatile oil are lost. Although the present method of procedure might give an arbitrary result, knowing the main source of error, it is believed that the method is not altogether satisfactory. The associate referee has tried by steam distillation to determine the amount of volatile oil in anise and caraway, and drugs of similar nature. The steam is passed through a certain amount of the crushed material and carries with it the ethereal oil, which is collected, salted out, and extracted from solution by means of an organic solyent such as ether, petroleum ether, or carbon tetrachlorid. It may be mentioned here that in some instances the addition of carbon tetrachlorid to the mixture before distillation proved to be successful in speeding the removal of the ethereal oil from the material. The solution of the volatile oil is con- centrated on a steam bath to a small quantity, usually to not less than Determination of volatile oils by spontaneous evaporation of the ethereal AMOUNT OF : NUMBER SUBSTANCE SUBSTANCE WEIGHT IMMEDIATELY TAKEN AFTER EVAPORATION OF ETHER gram gram per cent Dy Ua geMOl se ets As oe ee 0.1676 0.1824 108.8 2 iOilroficloves=s 10 Seton Ba eee eee ee 0.1492 0.1556 104.3 3 | |Oullofpumenta~ 222 s2 2a ee seen se 0.1394 0.1434 102.9 Ase” Oilltofthaysviesste e352) 3 le ee oe 0.1456 0.1370 94.1 Dim wCannamicraldehyde te =e aes 0.1585 0.1585 100.0 6a | Oukofcassiag#: = <5 Geese. eee 0.1908 0.2040 106.9 7. | sOuvof@eylon cmnamone] 222 seae ea oe 0.1700 0.1636 96.2 8" |*Oiltoftanise 3 sae t= oe ee LER eee 0.1574 0.1622 103.1 9) | :Oillofcaraway=!2:2<2- Lxseete ese = 8 0.6262 0.5242 83.7 LOW! Garvont; = aS oe te ae 0.1628 0.1648 101.2 Ak” OMofigingers2 2-2 Se ee ee 0.1784 0.1664 93.3 12s ROiliofspeppermin tas = Se eee 0.1440 0.1274 88.5 13%" Menthole ae Soe BT ee eee ae 0.1400 0.1330 95.0 1:4 i)) | (Camphor? S59 Se os. ek eee 0.3912 0.3836 98.0 DSi i ehvmole eee es hn ee 0.1148 0.1224 106.6 16h) || (Oiltofeucalyptuse)—- =e ee 0.1346 0.1216 90.3 | Onl ofturpentines <2 =" so See 0.2254 0.1689 74.9 18) ‘Oilloflemons= ses. 55 See ee 0.1540 0.1278 83.0 19|"Benzaldehyde2=22) 5. sk ee eee eee ee 0.2158 0.2003 92.8 10 cc. If evaporated down further, one can detect by the odor, or by a rise in the boiling point, that some volatile oil escapes with the organic solvent. The associate referee has tried to decrease the escape of volatile oil by blowing the solvent away with a gentle blast and drying the remaining ethereal oil completely for a few hours in a vacuum desiccator. The product thus obtained is undoubtedly the volatile oil without the 1920] VIEHOVER: MEDICINAL PLANTS 385 addition of fatty, waxy, or coloring matter, which is obtained in the indirect ether extract method. Thus far the associate referee has not obtained concordant results between the volatile ether extract and the oil obtained by distillation. Usually the amount of oil found by direct distillation was smaller than by the indirect method. This observation is confirmed by Hortvet', who obtained the amount of volatile oil in cloves by direct distillation with steam and indirectly as volatile ether extract. To show the loss of volatile oil through spontaneous vaporization of the ether and the drying of the ether extract in vacuum for an extended time, the following table is included. It is taken from Reich's work?, and shows convincingly that the loss of volatile oil always occurs, and that the amount lost depends on the nature of the volatile oil. solution and subsequent drying of the residue in vacuum. II Ill IV Vi WEIGHT AFTER WEIGHT AFTER WEIGHT AFTER WEIGHT AFTER STANDING 30 MINUTES STANDING 2 HOURS STANDING 24 HOURS STANDING 48 HOURS IN DESICCATOR IN DESICCATOR IN DESICCATOR IN DESICCATOR | gram per cent gram per cent gram per cent gram per cent 0.1746 104.2 0.1721 102.7 0.1661 99.1 0.1610 96.6 0.1486 99.6 0.1450 97.2 0.1420 95.2 0.1386 92.8 0.1394 100.0 0.1356 97.3 0.1314 94.3 0.1282 91.9 0.1290 88.6 0.1240 85.2 0.1034 71.0 0.0944 64.9 0.1585 100.0 0.1577 99.5 0.1481 93.4 0.1451 91.6 0.1940 101.7 0.1908 100.0 0.1840 96.4 0.1810 94.8 0.1600 94.1 0.1564 92.0 0.1456 85.7 0.1420 83.5 0.1554 98.7 0.1549 98.4 0.1524 96.8 0.1458 92.6 0.5197 83.0 0.5162 $2.4 0.4874 77.8 0.4266 68.1 0.1568 96.3 0.1548 95.1 0.1478 90.8 0.1348 82.9 0.1460 81.8 0.1649 92.4 0.1622 90.9 0.1498 83. 0.0844 58.6 th) 0.1208 83.8 0.1194 82.9 | 0.0964 66.9 | 0.1244 88.9 | 0.1184 84.6 0.1084 77.4 4 4 0.1264 90.3 0.3630 92.8 0.3472 88.8 | 0.2832 72. 0.2316 59.2 0.1100 95.8 0.1070 93.2 0.1050 91. 0.1060 92.3 0.1060 78.8 0.0988 73.4 0.0478 35.5 0.0298 22.1 0.1558 69.1 0.1474 65.4 0.0734 32.5 0.0640 28.3 0.1230 79.9 0.1196 77.7 0.0716 46.5 0.0434 28.2 0.1758 81.2 0.1638 75.9 0.1158 53.7 0.0958 44.4 To overcome the obvious errors through evaporation, the oil may be distilled directly and then separated mechanically and the volume de- termined. From this, together with the specific gravity, the actual 1 J. Assoc. Official Agr. Chemists, 1915, i: 154. ? Z. Nahr. Genussm., 1908, 18: 500. 386 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 percentage by weight may be calculated. Work has already been done along this line. Mr. Darling submitted the following: In regard to determination of the volatile oil content of drugs and spices, it is believed that the following method will give accurate results with volatile oil content of 0.5 per cent and upwards, i. e., aniseed, fennel, etc. The details were worked out in col- laboration with Mr. Elgar O. Eaton. Mix 10-25 grams of crushed seed with about an equal weight of sand. Place a pledget of cotton at the bottom of the inner tube of the Hortvet distilling apparatus. Introduce the prepared sample, tapping the tube gently, and cover with a pledget of cotton. Distil rapidly, using a condenser of the worm type. Collect the distillate in a flask having a graduated neck. When the receiving flask is nearly full, add sufficient pure salt to saturate the lower layer and read the volume of oil. In case the oil collects in small drops on the neck of the flask, it may be washed down with a little ether, and the ether removed by immersing the flask in the steam bath for a short time. The method is very rapid. With 10 grams of seed, practically all the oil comes over in the first 100 mils. It would conduce to greater accuracy if 25 grams of the sample were used, and a receiving flask of 200 mils capacity with neck graduated to 0.05 mils. We have been obliged to use the cassia flask which holds only about 100 mils. In a conference the associate referee suggested some modifications. The results of our attempts, as outlined, will be discussed later. PART III. A considerable number of crude drugs or spices has been offered for entry in this country in an adulterated or otherwise objectionable con- dition. The work on the identification of new substitutes has been especially difficult on account of the frequent lack of information or authentic samples. Striking examples of substitution or adulteration were the importation of Spanish digitalis (Digitalis thapsi), a non-official species, neither chemi- cally nor pharmacologically known; the importation of mustard sub- stitutes, such as Chinese colza, and Indian tori, for genuine mustards; of bitter fennel (Foeniculum piperitum) for the usual product, Foeniculum capillaceum. Samples submitted as ipecac were found to be obtained from other species not containing any alkaloid. Especially serious were the two following findings: Senna leaves containing 20 per cent or more of Tephrosia leaves, used as fish poison, yielding the bitter glucoside tephrosin; marjoram leaves with about 10 per cent of Coriaria myrti- folia leaves, which contain a poisonous substance acting similarly to picrotoxin. RECOMMENDATIONS. It is recommended (1) That work on the method for the determination of volatile oil of mustards and mustard substitutes be continued. (2) That work on the determination of ethereal oil in drugs and spices be continued. a 1920) CHESNUT: REPORT ON PAPAIN 387 REPORT ON PAPAIN. By V. K. Cuesnut (Bureau of Chemistry, Washington, D. C.), Associate Referee on Pepsin and Papain. The work on pepsin and papain was limited to investigations with papaya latex, mostly collected personally by the associate referee from papaya (Carica papaya) trees under cultivation chiefly at the Sub- tropical Experimert Station of the Office of Foreign Seed and Plant Introduction, at Miami, Florida. Dr. David G. Fairchild, in charge of the plant introduction work of the Department of Agriculture, generously consented to let the Bureau of Chemistry take samples from a portion of the fruits on each of all the varieties introduced there from all parts of the tropics, and also placed the Sub-tropical Laboratory at our dis- posal for drying the material. Forty-eight different samples were pro- cured and dried as rapidly as possible over calcium chlorid in a box, through which air, heated by electricity to about 50°C., ascended. These samples, together with a composite, slowly sun-dried specimen collected at Miami by Dr. H. H. Rusby, nineteen procured from Pro- fessor J. E. Higgins, horticulturist of the Hawaiian Experiment Station, a few other genuine samples obtained by the associate referee from fruit grown under glass at Washington, and various commercial brands and imports furnished the material from which the data included in this report were obtained. Papaya latex and papain, the partially purified product, are greatly misunderstood commodities. No attempt will be made to define either for this paper deals more particularly with the assay of dried latex. It must be borne in mind that “‘papaya latex” is not “papaya juice’’, as it is often erroneously called. [tis not obtained from the crushed fruit, which contains considerable sugar and little or no proteolytic enzym, but from the milky sap which gushes chiefly from the rind of the unripe fruit, and also in very small quantity from the surface of leaves and stems when punctured. It contains little, if any, sugar. It was found on observation that the character of this latex varied a great deal, de- pending upon the ripeness and size of the fruit, whether collected at once when punctured, or after a few minutes, and especially if gathered an hour or so after bleeding, or after a drizzling rain. In most cases the latex, gathered without contact with iron, was nearly white, but occasion- ally it was brownish or nearly black when drawn, and twice no latex whatever was obtained on puncturing the unripe fruit. The ripe fruit yields a small quantity of a viscous solution which is not milky and is nearly devoid of proteolytic activity. The latex of young fruits coagu- lates very completely and very quickly, that from fully developed un- ripe fruits coagulates less completely and much more slowly. The latex 388 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 which gradually exudes and dries upon fruit, from which one portion of latex has been collected, is generally of a slightly brownish color and a little gummy. The chief quantity of latex was obtained in every case from the freshly punctured rind of fully developed but unripe fruits, and this should most fully represent the commercial article to which a definition of papaya latex should apply. Still, the other forms of the latex can not be excluded until further field and laboratory studies are made of them. In general, however, it may be said that good papaya latex is always easily friable between the fingers and possesses little or no offensive odor. All of the writer’s samples were easily powdered and free from offensive odor, but nearly all were selected to represent chiefly fruit of medium size ranging down in greater or less number in different trees to the very youngest fruits. The largest fruits, though unripe, were left to mature. The writer was not then aware that the latex from young fruits was not very active, and wished particularly to study the latex of different horticultural varieties. His samples are, therefore, not strictly comparable for variety tests. The easy friability of the gen- uine latex permits one readily to distinguish the genuine from many of the hard, horny samples recently imported into the United States. These contained such a large quantity, often over half, of rice starch or bread stuffs, etc., that the importers, in order to have them admitted into the country, were compelled to relabel them in various ways so as plainly to indicate their approximate content of starch. This form of adulteration is the most common, but fortunately it is the most easily handled, for it has been found that there is practically no starch or sugar in the genuine latex. Papain is also sometimes vile smelling, indicating autolysis or bacterial or fungus decay due to slow and perhaps dangerously insanitary drying!. Such samples are of low activity, but occasionally samples are received, especially in interstate commerce, that are considerably more active than are those usually imported. In some of these cases the product was found to be adulter- ated with milk sugar and pepsin. Various methods have been suggested for the assay of papain, but few appear to have been based upon genuine samples, and none take proper cognizance of possible adulteration with pepsin, trypsin, and other lesser known proteoclastic and peptoclastic enzyms. It was found that several of the methods proposed did not exclude adulteration with pepsin and others did not exclude adulteration with trypsin or erepsin. The ordinary alkaline? method by digesting 8 grams of egg or fibrin in 100 cc. of water made slightly alkaline, upon which the commercial standard of 1 to 80 was adopted for papaya latex, was found to work quite as well with ‘TL. Huybertsz. Chemist and Druggist, 1916, 88: No. 1897, 51. 2 Real-Enzyklopiidie der gesamten Pharmazie. Zweite Aufl., 1908, 10: 4. 1920] CHESNUT: REPORT ON PAPAIN 389 trypsin, only 100 to 200 mg. of one of the best brands being required for the purpose. Trypsin adulteration not excluded under this method would perhaps be unprofitable even at three or four dollars per pound for the latex on account of the high price of trypsin; but if, for example, the Shelley! modification of S6rensen’s method of assay were accepted as official for papaya latex without qualifying tests, it would permit very gross adulteration with pancreatin or trypsin, because, as the author shows, good pancreatin can liberate from casein eight times as much amino acid as papaya latex, and do the work in only one-quarter of the time. It is, therefore, very evident that it is important first to decide upon what should be measured, and then to make sure by quali- tative tests that other enzyms do not furnish the measurement observed. Adulteration with starch can be detected quickly by its reaction with iodin, but it is manifestly impossible in a product which differs in its physical properties so widely as does this latex for one easily to detect all possible forms of adulteration. Papaya latex should, therefore, be purchased wholly upon the basis of the extent of its characteristic proteolytic action. What is the characteristic activity of papain? What is the power which we should measure, and upon what protein, if any, is its action most characteristic? Is its chief value in internal use, in the prepara- tion of meat extracts”, or the clarification of beer*? Papain can split up one and the same protein in a variety of ways in varying periods of time according to the temperature, the hydrogen-ion concentration and the presence or absence of other substances. As shown by Mendel and Blood‘ and verified by the writer, this is especially the case with prussic acid. It exerts a strong influence in shortening the time, as well as in increasing the depth, of cleavage. The reason for this is still unknown. Several such problems are at present unsolved, so we must be content with measuring some one characteristic action which may be definitely expressed in mathematical terms. A comprehensive review of recent literature indicated that uncoagu- lated casein is the most generally satisfactory protein for comparing the action of various proteolytic enzyms at various hydrogen-ion concen- trations. Its composition, especially its acid value when purified by the Hammarsten method, is nearly constant, and it is, therefore, an excellent material with which to make up solutions to a fairly definite hydrogen- ion concentration from a standard stock of casein by dissolving in stand- ard sodium hydroxid and adding standard hydrochloric acid. It is also generally available at a fair price,and may be made quite readily in the 1 Analyst, 1914, 39: 170. ?Barral. J. pharm. chimie 1905, 22: 395 2 Letters Patent. 1911, Nos. 995820 and 995824. 4 J. Biol. Chem., 1910, 8: 177. 390 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 laboratory. Its chemistry is as well known as is that of most of the proteins. Furthermore, it is perhaps the most representative protein food and contains considerable amounts of the lysine and tryptophane groups necessary for growth. It has been shown that papain has a quick digestive action on casein similar to that of trypsin and erepsin. Casein was especially recommended by Hedin! for use in assaying trypsin. Scheermesser? states that casein is good for pepsin assay, but not for trypsin. Neun* concludes that casein is good for the qualitative testing of both pepsin and trypsin and is much better than egg for this purpose. She also favors it somewhat for assay work but was unable to get strictly quantitative results. Bogdandy*, Shelley and Pratt® all used casein in their work. A very great variety of modern assay methods, comparative, gravi- metric, chemical, involving the determination of amino acid nitrogen and its ratio to total nitrogen, bromin absorption values, ete., and physico-chemical methods, involving the use of the polariscope, colori- meter, viscosimeter, refractometer, etc., were open for consideration in connection with the use of casein. The polariscope type of assay to which Fischer and Abderhalden® have paid much attention, especially in connection with their enzym work with the polypeptids, and which, moreover, has been worked out for pepsin in detail by Bogdandy, was finally selected as the most expeditious and best suitable in every way for the numerous assays involved. Assays with a gram of casein, as shown by Pratt, require only 10 mg. of the latex. This was a very valuable con- sideration in this work, for there were only a few hundred milligrams of some of the samples of latex. The possibility, in the use of such small quantities, in favoring breeding investigations with a view to improving the quality and quantity of papaya latex produced should be considered very especially in connection with this method. It is more reliable than gravimetric methods for, aside from the mere solution effect, it has been noticed that in papain digestion there is a certain amount of cleavage into simpler forms, including the more or less complex polypeptids and perhaps amino acids which can not be easily measured by the balance. The same objection applies to the alkaline? assay with fibrin or white of egg—it measures only the solution effect. The chief objection to this method, however, is the main objection to all of the methods thus far proposed—the indefiniteness of the hydrogen-ion concentration of the 1 J. Physiol., 1905, 32: 468. 2 Apoth. Zig., 1913, 28: 752. 7 3D. E. Neun. An Examination of Certain Methods for the Study of Proteolytic \ction. Dissert. Columbia Uniy., 1915, p. 39. 4 Z. physiol. Chem., 1913, 84: 18. 5 Philippine J. Sci., 1915, 10: 1. 6 Uber das Verhalten verschiedener Polypeptide gegen Pankreassaft und Magensaft— in Unterss. itiber Aminosiiuren Polypeptide u. Proteine. 1899-1906. p. 595. 7 Real-Enzyklopiidie der gesamten Pharmazie. Zweite Aufl. 1908, 10: 4. 1920] CHESNUT: REPORT ON PAPAIN 391 substrats used. This investigation has shown that considerable care must be taken, not so much to the acidity as to the hydrogen-ion concentration of the substrat. Papain has a comparatively narrow zone of activity. The method under consideration does not establish any definite amount of alkali to be added to the fibrin or egg and, therefore, no exact or even comparative measurement of activity is possible with it. Indeed, the method has, through this neglect, led to at least one re- port of total inactivity, whereas, as a matter of fact, the sample in question was found by the associate referee’s method to be an excellent latex with no enzym adulterant. Attention has been called to the fact that the Shelley method would permit gross adulteration with trypsin. It is believed that this is due largely to the unfavorable alkalinity of the casein solution used. This criticism may apply to Neun’s work, but in ‘her polariscope work she used too large an amount of some of the enzyms, and should not have expected comparative results with limited quantities of casein. Pratt did considerable work to show how narrow is the zone of papain activity, but adopted in his report the acidity of some unnamed commercial brand of sweetened condensed milk. It is conceivable that this acidity may vary considerably with different brands, especially in any that might be made of milk of questionable age. Perhaps ordinary casein is open to the same criticism, but the acidity of the uncoagulated casein, prepared by the Hammarsten process, has a fairly definite acid value. From 8.3 to 8.7 cc. of N/10 sodium hydroxid are required in different samples to make 1 gram neutral to phenolphthalein. Three 1 pound samples of Hammarsten’s casein were secured for this investi- gation and mixed intimately, by the use of a sieve, and the purity of the mixture determined from data given by Abderhalden'. In weighing out the casein for digestion experiments, due allowance was made for the moisture present by calculating the protein to the anhydrous basis. Standard carbon dioxid-free sodium hydroxid was pre- pared of N/5 strength by the Morey? method, using pure sublimed benzoic acid furnished by the Bureau of Standards, and this alkali was used in preparing standard hydrochloric acid. Sixty cc. of N/5 sodium hydroxid were added to 12 grams of casein previously thoroughly shaken in a 300 cc. flask with about 100-150 ce. of water, and after vigorously shaking the mixture for about 30 minutes until the casein was dissolved, the flask was filled to the mark with water. Twenty-five cc. of this alkaline casein solution, to which 20 cc. of water were added, had a hydrogen-ion concentration of P,,=10~°*°, a point determined approximately by the substrat turning very light blue with thymolphthalein. This figure was determined by Mr. G. H. Mains of 1 Bio. chem. Handlexikon, 1910-11, 4: 105. ?U.S. Bur. Standards Bull. 8: 643. 392 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 the Physical Chemistry Laboratory of the Bureau of Chemistry, and is the basal solution from which all substrats were derived by addition to separate 25 ec. portions in 100 cc. measuring flasks of varying amounts of water and of the standard acid or alkali, leaving a final volume of 45 cc. before the addition of the constant volume of 5 cc. adopted for the papain and other digestion solutions. The hydrogen-ion concentra- tions of all these various 45 cc. substrats have not yet been determined, but the solution found by the associate referee to be optimal for papain, the substitution in the basal solution of 2.5 ec. of N/5 hydrochloric acid for 2.5 cc. of water, was found by Mr. Mains to be P,=10~**!. The extent of activity was found to vary only a little when from 2 to 4 ce. of the N/5 hydrochloric acid were substituted for the same quantity of water. The activity drops off very rapidly if less than 2 cc. are added; less rapidly if more than 4 cc. are added. The substitution of 5 ec. of the acid for 5 cc. of water causes strong coagulation of the casein, but a 12.5 cc. substitution gives a clear solution which is about optimal for pepsin. The optimal zone for papain may be fairly well determined with indicators. The 45 cc. solution to which the very slightly acid papain solution is added should be basic to methyl red and phenolsul- phonephthalein and acid to paranitrophenol and phenolphthalein. The substrats in the 100 cc. measuring flasks referred to above were placed in a water bath held constant at 37.5° C. for about 30 minutes while the solution of the latex or enzym under examination was being prepared. Water and 1 per cent sodium chlorid solution, both with and without toluene, were tried as solvents, but 1 per cent sodium chlorid solution without toluene was found best. One hundred and fifty mg. were ample material for acidity determinations and for eleven assays. This was placed in 75 cc. of 1 per cent sodium chlorid and digested with occasional agitation at room temperature for 30 minutes. It was then filtered rapidly through a folded filter, transferred to a burette and added in 5 ce. portions, representing 10 mg. of latex, to the warmed substrats, which were then replaced in the bath and digested exactly 30 minutes. They were then removed and to each was added immediately 30 cc. of the Bogdandy precipitating mixture (consisting of 132 grams of an- hydrous sodium sulphate, 100 grams of magnesium sulphate and 200 ce. of 95 per cent alcohol diluted up to 2000 cc. with water), and 5 ce. of N/1 hydrochloric acid. The mixture and the acid were previously measured out in flasks so that they could be added quickly. The flasks were cooled to 20°C. and water added to the 100 ce. mark. The flasks were well shaken, the contents filtered and readings made with the filtrates at 17°C. in 200 mm. tubes ina S. & H. half-shadow polariscope, by means of which the angular rotation could be read accurately to within 0.01° or 0.02°. No difficulty was experienced in obtaining clear filtrates 1920| CHESNUT: REPORT ON PAPAIN 393 except in the case of the most active samples. These had to be filtered repeatedly through double filters. Many samples of latex prepared in various ways from all available varieties of fruit in all stages of growth were tested at from 3 to 11 different hydrogen-ion concentrations, but the maximum effect was invariably obtained when the enzym was added to the 45 cc. substrat, the hydrogen-ion concentration of which was P,=10~°°'. The samples thus tested included, among others, latexes produced naturally from one year old trees in Florida, Honolulu, and Montserrat, and under glass at Washington, and one from the very small fruit of a tree several years old; those representing rapidly and slowly coagulating forms and coagulation residues, some of recent collection and some two to five years old, and one exposed to the air of the laboratory for nearly a year; those ‘dried slowly and rapidly in the sun, at a moderate heat in a dryer and at a temperature a little below the boiling point of water. When the same samples were tested at the optimal hydrogen-ion concentration for pepsin, little or no activity was shown, as was also the case when tested at the optimum for trypsin. Very considerable activity was shown, however, both by trypsin and by papaya latex at a concentra- tion about half way between their optimal points. ie IN, Fic. 1. Uncorrected curves showing influence of hydrogen-ion concentration upon 10 mg. enzym diges- tions of 1 gram of casein: 1. Correction curve; 2. U.S. P. scale pepsin; 3. Best papaya latex, Inv. No. 4934: 4. One of the best brands of trypsin. Thirty minute digestions at 37°C. 394 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 Figure 1 furnishes a detailed sketch of the work of these three enzyms and gives a correction curve, No. 1, showing the amount of rotation found when 5 cc. portions of water are substituted for the enzym solu- tion and the assays carried out in the same manner as the enzym assays. Five ce. of an aqueous solution of 10 mg. of the enzym were added in each case to substrats containing exactly 1 gram of anhydrous casein, and the digestion carried on 30 minutes at 37.5°C. The abscisse oO SO°_ 20° .50° 40° 50° 60° 70° .GO° .9F 400° 440° 2D AGO? 14 O°LSO” A G hy \ VULLIG2EVAS OF FLAN LMVER (VU NO. FORE Q go Fic. 2. Corrected curve showing negative-rotation effect of different quantities of latex, Inv. No. 4934, upon | gram of casein at the optimal hydrogen-ion concentration. Thirty minute digestions at 37°C. represent the number of cc. of hydrochloric acid, or of sodium hydroxid, added to the basal solution to bring the volume up to 45 ce. after addi- tion of water; the ordinates represent the negative polariscopic reading at 17°C. of the filtrates from the digestion mixtures after precipitation by the Bogdandy mixture and dilution to 100 cc. The actual points of observation are indicated by dots. Scale pepsin of U. S. P. value was used for Curve No. 2, and one of the best commercial brands of trypsin for Curve No. 4. The points for Curve No. 3 were furnished by Florida latex, Investigation No. 4934, obtained by drying the juice at about 50°C. over calcium chlorid. Tt is characteristic of all of the many samples examined that the highest point of all the curves was invariably at the hydrogen-ion concentration here shown as the optimum and that the curves descend very rapidly on both the acid and alkaline sides. This 1920] CHESNUT: REPORT ON PAPAIN 395 latex was the most active one examined, but the best Hawaiian sample was little, if any, inferior. Some samples were so nearly worthless that the maximum rotation found was very low, as was the case with the five year old Hawaiian sample, Informal X, I. S. No. 8019-K, the read- ing for which was —0.13°. The average reading found in the case of the Florida samples was 0.54°; that for the eighteen Hawaiian samples, 0.80°. These figures, however, can not in any way be regarded as show- ing the comparative activity of the latex from the two places. Had the latex of selective individual fruits been compared, very much higher averages would have been found in both cases. Variety tests for com- parative yield and activity must be carried out only with latex from fully developed but unripe fruit. In Figure 2 the ordinates show the negative rotation at 17°C. as above, but in this case the abscisse represent the quantity in milligrams of latex No. 4934 added to the ten substrats, which were in every case of the optimal hydrogen-ion concentration. Nine-hundredths of a degree was subtracted in each case from the observed reading, so that comparison may be made at once with the change in deviation furnished by any 10 mg. sample of latex and a close estimate be made of its comparative strength. If sample No. 4934 were rated at 100 per cent, a net change in rotation of 1.07° caused by any 10 mg. sample would indicate that the sample is of 50 per cent strength. The curve found is not an altogether perfect one, but it is believed to be very serviceable for papaya latex work. A study of the two fig- ures will show how easily enzym adulteration may be detected and the papain value of any spurious sample of latex may be determined from a few polarimetric readings. For the detection of possible adulteration with trypsin, the polarimetric results must be supplemented with quali- tative tests. No gravimetric determinations were made to show the comparative amount of undigested protein after 30 minutes’ digestion at the optimal acidity for different samples, but simple inspection in a great many cases showed that there was a general parallelism, especially in the weaker latexes, with the results of polarimetric observation. But with the full 10 mg. amounts of the best samples, the parallelism ceased. Consider- ably greater deviation was noted in 8 to 10 mg. quantities after apparent solution had nearly ended. _ The measurement of enzymic activity will always be somewhat intangible until we have a more complete knowledge of the structure of the protein molecule. If we conceive it to be made up of a complex aggregation of polypeptid linkages, as it apparently seems to be, it would undoubtedly be better to measure the extent of proteolytic action by 396 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 the ratio of amino nitrogen found at any one time, not to the total nitrogen, but to the total amino nitrogen found upon complete hydrolysis. Sérensen' was the first to propose the formalin method of measuring this action, but several methods of making the same determination have more recently been advanced. Of all these, the Sérensen and the Van Slyke? methods were tested out to a limited extent. Dr. J. F. Brewster of the Bureau of Chemistry kindly made a Van Slyke determination for the writer, but difficulty was encountered in the profuse frothing of the reaction. It requires very much more time than the polariscopic method and was consequently abandoned. The Sdrensen method was not adopted for the same reason. The method here given is a comparatively simple one, very well adapted for research and seems well adapted for the assay of papain, especially where adulteration with other enzyms is suspected. It is hardly possible for any proteolytic enzym to show no activity in sub- strats with the eleven hydrogen-ion concentrations selected. In prac- tice the field can be amply covered by three assays made at the optimum for pepsin, trypsin and papain, respectively. Pepsin is the only enzym adulterant which has been reported in papaya latex and in papain, and the method of itself serves admirably to indicate its presence. If further tests be needed to detect pepsin or trypsin, the following will be of service. The presence of trypsin may readily be detected by the very great ease with which it breaks up glycyl-I-tyrosine (“‘peptone Roche’) with the formation of the very characteristic and easily recognizable bundles of needle-shaped crystals of tyrosine’. To 0.2 gram of this di-peptid add 5 cc. of a freshly prepared and filtered solution made by digesting 5 grams of the latex 30 minutes at 37°C. in 25 cc. of water to which 1 ec. of N/1 potassium hydroxid has been added. Add 2 drops of toluene, stopper loosely and digest at 37°C. If trypsin or certain peptoclastic enzyms are present, the crystals of tyrosine will gen- erally appear after 8 to 24 hours’ digestion. The high price of trypsin will militate against its use to increase the activity of any but the poorer grades of papain, for which. a high price is demanded. Papain also separates tyrosine from glycyl-l-tyrosine, but only in very minute quantities. The rapid, 80-90°C., digestion’ by papain of a solution of raw egg albumen in water slightly acidulated with acetic acid is shared, so far as known, only by bromelin, which is probably not used as an adulterant of papain. The digestion mixture is brought to the boiling temperature in a test tube within a couple of minutes and the biuret reaction is then 1 Biochem. Z., 1907, 7: 45. * Proc. Soc. Exp. Biol. Med., 1910, 7: 46. 3 L. Michaelis. In Abderhalden’s Handb. der Biochem. Arb. Methoden. 1910, 3: 21. * Soc. de Biol., 1906, 60: 309. 1920] CHESNUT: REPORT ON PAPAIN 397 applied to the filtrate. A strong pink color is produced if papain is present. It must be remembered, however, that unless especially pre- pared, pepsin contains peptones. Another high temperature test of great historic and practical interest is a modification of a very simple experi- ment originally made by Roy}, one of the first investigators of the latex. To a single 10 gram piece of tough beef immersed in 25 cc. of water, held at 75-80°C., add a half gram of powdered latex. The beef is reduced almost to soup within 15 minutes by a good latex, but is unchanged by a like weight of either pepsin or trypsin. The meeting adjourned at 4.15 p. m. for the day. 1 J. Med. Chirug. Pharmacologie, 1874, 59: 252. SECOND DAY. TUESDA Y—MORNING SESSION. REPORT ON FOOD ADULTERATION. By Jutrus Hortvet (Dairy and Food Department, St. Paul, Minn.), Referee. It is a satisfaction to note among the referee reports of the past two years not only a widening of our field of study of analytical methods of procedure, but also a reaching out into subjects relating more than here- tofore to basic principles underlying the science of modern chemistry. We are not unaware of the advances which have been made from the more or less purely empirical methods which constituted in large measure our great reliance not many years ago. As a conspicuous case in point, the application of methods of study following mainly theoretical lines has resulted in an entire rearrangement of the chapter on coloring matters in our revision of official methods of analysis. Take also as illustrations the referee reports on fruit products, dealing with a collaborative study of methods for the quantitative determination of the common fruit acids; investigations taken up on baking chemicals, involving not only ordinary methods of analytical procedure, but also suggesting the application of principles of physical chemistry hitherto seldom enlisted in our attempts to overcome difficulties; and the plan of work demanded as a result of recent advances in the technology of edible oils and fats. There is no doubting the fact, and this point must be kept ever plainly before us, that we have recently arrived at a critical turning point or at a point of greatly increased impetus in the forward movement of that branch of applied chemistry dealing with the analysis of food products and methods of detecting adulteration. In line with this brief statement of facts intended to describe, in a general way at least, the conditions under which we are laboring, it seems necessary to speak a word relative to the subject titles which have been announced in the programs of our meetings for two or three years past. Considering the state to which we have actually advanced in our work, it is clearly apparent, in a number of instances at least, that our present subject titles are seriously in need of revision, and that there is also doubtless a call for the addition to our program of a number of subjects hitherto not taken up in connection with our collaborative work. In the first place, a clearer distinction should be made in the matter of subject headings between those lines of investigation which 398 1920] HORTVET: FOOD ADULTERATION 399 are of special interest to those chemists engaged in experiment station work, or work relating chiefly or wholly to agriculture or dairying, and those lines of investigation which are related entirely to the general subject of food analysis and methods designed for the purpose of detect- ing food adulteration. On examining the program we find that the subject, dairy products, appears in two places and is handled by two separate referees. A distinction in title should be made between these two subjects, inasmuch as it is obvious that the two lines of investigation have not a common purpose in view. The work on dairy products, which relates primarily to experiment station work or has to do with special problems not in any way essentially connected with the work relating to food adultera- tion, may be continued under its present title, viz., dairy products; and by way of differentiating in a matter where a real distinction certainly exists, it is therefore suggested that the present title, dairy products. under food adulteration, be changed to read ‘Milk and Milk Products”. In the same manner, we already have a fair distinction between the subject, sugar, which appears on the general program of the first day, and the subject, saccharine products, which occurs in the group under the heading “‘Food Adulteration’’. This plan of differentiating one class of work from the other is carried out apparently well enough in the present arrangement of program and no further suggestions on this particular point seem to be necessary. Under the general subject, food adulteration, however, there is clearly a call for further revision in the matter of subject titles. The associate referee on cocoa and cocoa products has very properly recommended and urged that this subject be changed so as to read ““Cacao Products”. A thorough examination into the origin and meaning of these terms will reveal the necessity for such a change. Also, along the line of broadening our work as well as in harmony with our plans to intensify, it is suggested that the title, flavoring extracts, be changed so as to read “Flavoring Extracts and Essential Oils”; that the title, spices, be broad- ened to read “Condiments and Spices”; that the title, baking powder, be changed to “Baking Powder and Baking Chemicals”; that the sub- ject, beer, be broadened so as to read ‘‘Malt Liquors and Beer’; and that the subject, wine, be expanded into the subject “Fermented Fruit Juices, Wines and Ciders”’. Other changes, though probably of minor importance, may be sug- gested, especially in view of the fact that food control work has recently been extended into a good many classes of products not heretofore seriously taken into consideration. We find ourselves in recent years dealing more and more with crude materials, the original natural products from which manufactured products are derived. Along this line it is 400 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 proposed that we expand the subject, colors, to “Colors and Commer- cial Dyestuffs”; cereal products to “Cereals and Cereal Products’’; fruit products to “Fruits and Fruit Products’; and that the subject, vegetables, be enlarged to “Vegetables and Vegetable Products’. Rela- tive to the subject, meat and fish, it is apparent that a fictitious distinc- tion has inadvertently been made between the meanings of two terms of common usage. Doubtless there may be individuals who make a certain distinction between meat and fish; nevertheless, in harmony with the purposes and plan of our work, it is recommended that this subject be changed so as to read ‘““Meat and Meat Products’. In the above list of revised titles the further recommendation is made that we include two additional subjects under the respective headings “Confectionery” and “Soft Drinks’. These recommendations are offered in the belief that they are in the main timely, and it is therefore urged that this matter be taken up for consideration by the appropriate com- mittee having in charge the recommendations of referees. REPORT ON COLORS!. By W. E. MatHewson (Bureau of Chemistry, Washington, D. C.), Associate Referee. Chapter XX VIII of the official methods?, which has not been changed for a number of years, was thoroughly revised by the Committee on Editing Methods of Analysis. It seemed that the tentative methods should be brought before the association for criticism this year, and with this in view copies of the draft were sent to the chemists collaborating on the color work. It was suggested to the collaborators that they make a few mixtures from food products and colors that chanced to be of special interest to them, try the methods on these, and report criticisms to the association. In addition, it was hoped that some advance might be made in the selection of methods for the natural coloring matters. However, little was attempted or accomplished in this field*. The statements of the collaborators follow. Mr. L. A. Salinger (U. 5. Food and Drug Inspection Station, U. S. Custom House, Savannah, Ga.), who made a very complete examination of the methods, has abstracted his full report. 1 Abstract. * 2U.S. Bur. Chem. Bull. 107, rev.: 190. " 31.8. Palmer and W. E. Thrun have recently published a paper (J. Ind. Eng. Chem., 1916, 8: 614) discussing the detection of natural and artificial coloring matters in oleo- margarine and butter. They conclude that it is not possible to distinguish added carrotin from that naturally occurring in the fat. 1920] MATHEWSON: REPORT ON COLORS 401 L. A. Salinger —The separation of the eight permitted coal tar colors by immiscible solvents was selected from the work outlined by the association. In this separation I used aqueous solutions and solutions of food products that were freed from alcohol by evaporation, adding one-half volume concentrated hydrochloric acid and extracting the color by amy! alcohol and washing out the color from the amyl alcohol as outlined in the procedure, using N/4 hydrochloric acid in the first washings. In order to make the behavior of the colors comparable, all the extractions and washings were made in the same way. The procedure was tried first on a mixture of the eight permitted colors; then combinations were made with different food products, including cordials, grapefruit, marmalade, etc. From all these extractions the colors were separated and were properly identified by the test on wool, matched against standard colors. One fractionation was sufficient, especially if the first and last washings of each group were rejected. In separating Naphthol Yellow S and Ponceau 3 R from each other, it was found that they were completely separated by adding salt to saturation. This is more com- plete than the ethyl acetate separation. The Naphthol Yellow S does not dye readily from this strong salt solution. It is best to take out the color with acid amyl alcohol. Then the wool will dye readily in the aqueous washing. Tartrazine, Indigo Carmin and Amaranth were easily separated from each other by means of sodium hydrosulphite. Other reagents were tried, but were not successful. In washing out Erythrosin from Orange I, it is best to use distilled water. Our tap water was alkaline enough to wash out the Erythrosin completely in two washings. My idea of the description of the colors produced by the various reagents of the eight permitted colors dyed on the cloth tallied very closely with that described in the pro- cedure. I should say the scheme outlined in the procedure will give satisfactory results with any person using ordinary care in following out the instructions. By refractionation and using definite amounts of solutions, reagents and number of times of washing, the groups can be completely separated and entirely freed from each other, especially if the first and last washing in each group is rejected. C. L. Black (U. S. Food and Drug Inspection Station, U. S. Appraiser’s Stores, Phila- delphia, Pa.).—I have made a very critical study of the methods submitted for the association work, and have also tested them in various ways in which I would expect defects to show up, and have found them quite satisfactory. I feel that the greatest mistake in these methods lies in failure to caution the analyst as to the separation of colors not being sharp and not to use the portions obtained by the final extractions with one strength acid in his identification of the color extracted with that acid. I made a number of individual tests and found that a mixture of the permitted colurs was separated and identified with ease, using the above caution. Then to a mixture of the permitted colors, with the exception of the blue and green, I added Martius Yellow, 8. & J. No. 3, and found no interference. Toa similar mixture Coccine, S. & J. No. 106, was added and appeared with the Amaranth, from which I know of no way to separate it; Orange II, when added, appears as a contamination of the Erythrosin. For some time the following test for Erythrosin has been used in this laboratory: A small amount of the dyestuff is obtained by stripping the dyed fiber and evaporating in a porcelain dish. A beaker (which will fit inside the porcelain dish) is then prepared containing ice-water and a lump of ice and on the bottom outside a drop of starch paste. Sulphuric acid is then added to the dyestuff in the dish, the beaker placed over it and the dish heated on a wire gauze, liberating iodin, which is shown on the drop of starch piste. On the whole, I feel that there is no method of separating dyes so satisfactory as by using a chart on the solubility of dyestuffs in various organic solvents when shaken with acids, etc. Then, by starting at the highest strength acids and coming on down the 402 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 list, the analyst can see so much more easily what dyes it is possible he may be extract- ing at any point. If such a chart could be made sufficiently complete to cover prac- tically any possible color encountered, the analyst would need nothing but it and his table of reactions of dyed fibers and a few specific chemical tests which are known for various dyestufis. No work on the natural colors was reported this year by collaborators. TARTRAZINE"’. By Aiwa M. Doyie? (Bureau of Chemistry, Washington, D. C.). Short reference was made to the reasons why an additional yellow color for foodstuffs was needed and to the action of the Department of Agriculture in issuing Food Inspection Decision 164 permitting the use of tartrazine. The components, methods of manufacture and con- stitution of the dye were briefly discussed and illustrated. As an aid in the understanding of the properties of the dye, brief reference was made to the uses of tartrazine and other pyrazolon dyes in the indus- tries, special emphasis being laid on the structure as it affects the tests for identification. A number of identification tests for tartrazine, on the dry dye, the solution 1 to 1000 in water and the dyed fiber, were described and compared with those applied to Naphthol Yellow S, mimeograph copies of some of the more distinctive tests being supplied. Methods of separa- tion of the color from the foods themselves, particularly paste products, were suggested and a number of illustrative samples shown. No report was made on saccharine products, as there was no associate referee on this subject. REPORT ON FRUIT PRODUCTS. By P. B. Dunpar’ and H. A. Lepper (Bureau of Chemistry, Washington, D. C.). In accordance with the recommendations made in the last report on fruit products, the work has been confined to studies of the uranyl acetate method for the determination of malic acid and of the Kunz modification’ of Stahre’s method® for the determination of citric acid. 1 Abstract. 2 Present address, 1365 Oak Street, Washington, D. C. 3 Associate referee. 4 Arch. Chem. Mikros., 1914, 7: 285; C. A., 1915, 9: 687. 5 Nordisk Tidskrift, 1895, 2: 141; Z. anal. Chem., 1897, 36: 195. 1920] DUNBAR: REPORT ON FRUIT PRODUCTS 403 MALIC ACID. Three procedures for the determination of malic acid, adapted to various types of fruit products, have been devised. These are based on the polariscopic method proposed by Dunbar and Bacon! and embody a number of the suggestions made by the previous associate referee on fruit products, Mr. H. C. Gore?. For advice and assistance in devising these procedures, the writers are much indebted to Messrs A. F. Seeker and R. E. Doolittle. The malic acid methods are as follows: 1 Method I. (For fruit juices and similar products which contain no tartaric acid and not over 15 per cent of sugars and in which the color does not interfere with polarization.) Filter the sample, if necessary to secure a solution which can be polarized readily, and polarize, using a 200 mm. tube if possible. If no free mineral acids are present, it is unnecessary to neutralize the sample before subjecting to the treatment which follows. If the sample contains free mineral acid, transfer a measured portion (75 cc. is a convenient volume) to a 100 cc. graduated flask; add enough standard alkali to neutralize the total acidity; dilute to the mark; mix well and filter. Transfer 25 cc. of the sample, or of the neutralized solution, to a flask graduated to 25 and 27.5 cc.; add about 2.5 grams of powdered uranyl acetate and shake vigorously at frequent intervals for 3 hours, keeping the mixture well protected from light. [f all of the uraayl acetate. dissolves, more should be added and a small amount should remain undissolved at the end of 3 hours. Dilute the solution to the 27.5 cc. mark with saturated uranyl acetate solution; mix well and filter, if necessary. Polarize, if possible, in a 200 mm. (or longer) tube. If the solution is too dark to polarize in a 200 mm. tube, a 100 or 50 mm. tube may be used. Multiply the reading by 1.1 to correct for dilution and, if a neutralized solution was used, make a further correction for that dilution. Multiply the algebraic difference in degrees Ventzke between the corrected readings, obtained before aad after treatment with uraayl acetate, calculated to the basis of a 200 mm. tube, by the factor, 0.036, to obtain the weight of malic acid in the sample in grams per 100 cc. Make all polarizations at room temperature with white light, taking care that all solutions are polarized at the same temperature. Make at least 6 readings in each case and take an average of these. In the case of dark colored fruit juices which can not be polarized readily, approxi- mately quantitative results may be obtained by adding to the solutions a few drops of bromin, shaking thoroughly and filtering just before polarization. Method IT. (Approximate determination for fruit juices and similar products containing no tartaric acid and more than 15 per cent of sugars.) 2 PREPARATION OF SOLUTION. Weigh out 25 grams of the sample and transfer to a 600 cc. beaker with a little 95% alcohol. Add alcohol, a little at a time, stirring the mixture well and warming, if U.S. Bur. Chem. Cir. 76. 2U.S. Bur. Chem. Bull. 162, 65; J. Assoc. Official Agr. Chemists, 1915, 1: 480. 404 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 necessary, to insure perfect solution of all alcohol-soluble substances, until 200 cc. have been added. Filter on a Biichner funnel, using suction, and thoroughly wash the precipitated pectins and insoluble matter with 95% alcohol. Disregard any slight turbidity which may appear in the filtrate when the washings are added. From the determination of total acidity, calculate the amount of N/4 barium hydroxid solution required nearly to neutralize the acidity in the 25 grams of sample taken. To the combined filtrate and washings in an Erlenmeyer flask, add the calculated quantity of barium hydroxid solution, stir until reaction is complete and then add 3-5 drops, or more if necessary, of an aqueous solution of barium acetate (50 grams in 100 ce.) to insure the presence of more than sufficient barium to combine with all the acid in the sample. Make up the volume of the mixture to about 375 cc. (not less) with alcohol, and reflux until the precipitate settles readily after being shaken. This may require 3-4 hours. Filter with suction and thoroughly wash the precipitate in the flask and on the paper with 95% alcohol. Transfer the portion on the filter quantitatively to the original flask, rinsing the paper with a jet of hot water for this purpose. Digest the precipitate with hot water, containing 2 grams of sodium sulphate in solution until the reaction is complete, and boil until the barium sulphate precipitate settles readily. Concentrate by evaporation, if necessary, and transfer quantitatively to a 100 ce. volumetric flask with a little hot water, cool, and make up to volume with water. Filter and treat portions of the filtrate as directed in 3. 3 DETERMINATION. Transfer 25 cc. of the solution, obtained as directed in 2, to a volumetric flask gradu- ated to 25 and 27.5 ce.; add about 2.5 grams of pulverized uranyl acetate and shake vigorously at frequent intervals for 3 hours, keeping the solution well protected from light. If all the uranyl acetate dissolves, more must be added in order that a small amount may remain undissolved at the end of 3 hours. Dilute the solution to the 27.5 cc. mark with a saturated uranyl acetate solution, mix well, filter if necessary, and polarize, using the same precautions as described in 1. Multiply the reading by 1.1 to correct for the dilution. Polarize another portion of the filtrate, obtained as directed under 2, which has not been treated with uranyl acetate. Multiply the algebraic difference in degrees Ventzke between the two readings, calculated to the basis of a 200 mm. tube, by the factor, 0.036, to obtain the weight of malic acid in grams per 100 cc. in the solution as obtained in 2. Method IIT. (Approximate determination for products containing tartaric acid.) 4 PREPARATION OF SOLUTION. Prepare the sample as directed under 2, up to the point of filtration and washing of the barium malate precipitate, then dry the precipitate thoroughly and transfer the portion on the filter quantitatively to the original flask, which has been previously dried, rinsing the paper with a jet of hot water for this purpose. Digest the precipitate with hot water, transfer quantitatively to a 100 cc. volumetric flask with a little hot water, cool, make up to volume with water and filter to remove undissolved barium tartrate. This amount of water is sufficient to dissolve barium malate up to amounts as large as approximately 0.9 gram. More than 100 cc. of water must be used when more than 0.9 gram of barium malate is present. The amount of barium tartrate dissolved by hot water is so small as to affect the polarization, after treatment with uranyl acetate, only to a slight extent. y 1920) DUNBAR: REPORT ON FRUIT PRODUCTS 405 5 DETERMINATION. Proceed as directed in 3, using the solution prepared as directed in 4. CITRIC ACID. The Kunz-Stahre method for the determination of citric acid is essentially that given in the report of last year, although a number of minor changes have been made. It differs little from the method originally published by Kunz except for the fact that the citric acid is first precipitated as barium citrate. Quantitative Determination. (Applicable in the presence of sugar and malic and tartaric acids.) 1 REAGENTS. (a) Barium hydrorid solution—Approximately N /4. (b) Barium acetate solution—Fifty grams of barium acetate in 100 cc. of water. (€) Dilute sulphuric acid—Equal volumes of sulphuric acid and water; also one volume of sulphuric acid and five of water. (d) Potassium bromid solution —Fifteen grams of potassium bromid in 40 ce. of water. A solution of sodium bromid (16 grams in 50 cc. of water) may be substituted for the potassium bromid solution. (€) Potassium permanganate solution —Five grams of potassium permanganate iu 100 cc. of water. (f) Ferrous sulphate solution —Twenty grams of ferrous sulphate in 100 cc. of water containing 1 cc. of concentrated sulphuric acid. 2 DETERMINATION. Proceed as described under ‘‘Malic Acid, 2,” to the point where the precipitated barium salts are washed and transferred to the original flask. Transfer the precipitate quantitatively from the filter to the flask with a jet of hot water; boil until alcohol can no longer be detected by odor and add enough dilute sulpharic acid (1 to 5) to precipitate all the barium originally added and to allow 2 cc. in excess. Evaporate by careful boiling to a volume of 60-70 cc., cool and add 5 ce. of freshly prepared satu- rated bromin water, or enough to show a distinct excess. Transfer quantitatively to a 100 ec. volumetric flask, with water, and dilute to the mark at standard temperature with water. Mix thoroughly, allow the precipitate to settle and filter through a dry paper. The precipitate may be separated by centrifugalizing and the supernatant liquid decanted, if necessary. Pipette an aliquot portion of the filtrate, containing _ not more than 400 mg. of citric acid (calculated from the total acidity of the sample), into a 300 ec. Erlenmeyer flask. The amount of the citric acid in the aliquot should, _ if possible, exceed 50 mg. Add 10 ce. of sulphuric acid (1 to 1) and 5 ce. of potassium _ bromid or sodium bromid solution, mix, warm the flask in a water bath to 48—50°C., and allow it to remain in the bath for 5 minutes. After removing from the bath, add 25 ce. of 5% potassium permanganate solution from a burette, in rapid drops, with frequent interruptions and constant, vigorous shakiag, care being taken that the tem- perature during oxidation does not exceed 55°C. Set the flask aside until the hydrated peroxid of manganese begins to settle. The supernatant liquid should be dark brown, showing an excess of permanganate (if an excess is not indicated, add more perman- 406 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. 1II, No. 3 ganate). Shake, again set aside to settle, and repeat this operation until the precipi- tate takes on a yellow color and most of it has dissolved. Finally, while the solution is still warm, remove the last undissolved portion of hydrated peroxid of manganese pre- cipitate and also the excess of bromin by adding drop by drop a clear, concentrated solution of ferrous sulphate. Allow the solution to cool with occasional shaking. If the operations have been properly carried out, a heavy white precipitate of penta- bromacetone is obtained which becomes crystalline on occasional shaking and, in this condition, is entirely insoluble in water. Allow the mixture to stand overnight, collect it by means of gentle suction in a porcelain Gooch crucible provided with a thin pad of asbestos, previously dried over sulphuric acid in a vacuum desiccator, wash with water slightly acidified with sulphuric acid and finally wash twice with water. Dry the precipitate to constant weight over sulphuric acid in a vacuum desiccator, pro- tected from strong light. The weight of pentabromacetone multiplied by the factor, 0.424, gives the equivalent weight of citric acid (H;C;H;O7). It sometimes happens that the pentabromacetone is first obtained in the form of oily droplets. These also become crystalline, on standing or on cooling, but are usnally discolored by negligible traces of manganese or iron. The accuracy of the result is not vitiated when this occurs. The above method may be applied directly to the sample under examination without previous precipitation of the citric acid as the barium salt when the amounts of sugars or other permanganate reducing substances are not excessive. In this case the deter- mination should begin with the addition of 2 cc. of sulphuric acid (1 to 5) and saturated bromin water. COLLABORATIVE WORK. In the original publication! the most favorable limits of concentration for the malic acid method are given as between 0.2 and 2.5 per cent. Method I for the determination of malic acid in fruit juices and similar products containing not over 15 per cent of sugars and no interfering colors, is practically identical with the original method which was made the subject of collaborative work by the previous associate referee, Mr. Gore®. A sweet cider containing 0.50 per cent of free malic acid as determined by titration was sent out by him to eight collaborators. The results reported varied from 0.468 to 0.531 per cent and averaged 0.59 per cent of malic acid. These results were so satisfactory that it was not considered necessary to give Method I a further trial in the — collaborative work this year. : The determination of malic acid in the presence of large amounts of f sugars, or of tartaric acid as in Malic Acid Methods II and III, presents } a) ee 2 difficulties not encountered in ordinary fruit juices. In order that these procedures might be given a severe test, it was thought advisable to prepare samples of jelly containing known amounts of malic as well as citric acid, and in some cases tartaric acid, for analysis by the collab- — orators. ; Numerous determinations of citric acid in aqueous solutions and in — fruit juices were tabulated in last year’s report and the results indicated — ‘U.S. Bur. Chem. Circ. 76: 2U.S. Bur. Chem. Bull. 162: 63. 1920} DUNBAR: REPORT ON FRUIT PRODUCTS 407 that the method could be applied to these products with reasonable accuracy. No determinations on products high in sugar had been made, however, and it was decided to apply the method to the same samples of jelly used for the collaborative study of the malic acid methods. TaBLe 1. Percentages of acids added to samples of jelly sent to collaborators. SAMPLE NUMBER MALIC ACID CITRIC ACID TARTARIC ACID per cent per cent per cenl 1 0.35 0.38 0.00 2 0.37 0.40 0.00 3 0.37 0.29 0.17 4 0.35 0.27 0.16 TABLE 2. Percentages of malic and citric acids found in samples of jelly sent to collaborators. Sampie No. 1. (Present: Malic acid, 0.35 per cent; citric acid, 0.38 per cent.) SamMpLe No. 2. (Present: Malic acid, 0.37 per cent; citric acid, 0.40 per cent.) ANALYST Raymond Hertwig, U. S. Food and Drug Inspec- tion Station, U. S. Ap- praiser’s Stores, San Fran- cisco, Cal P. L. Gowen, Bureau of Chemistry, Washington, L. F. Hoyt, Larkin Com- pany, Buffalo, N. Y. H. A. Lepper, Bureau of pacity: Washington, M. B. Porch, H. J. Heinz Company, Pittsburgh, Pa. G. W. Trainor, Armour and Company, Chicago, III. MALIC CITRIC ACID ACID per cent| per cent 0.29 | 0.32 0.29 | 0.32 0.36 | 0.30 0.34 | 0.29 Seed lee oA sy) 0.30 | 0.31 0.33 | 0.54 es 18029, Seen Os2e 0.25 | 0.32 0.24 | 0.32 ANALYST | F. D. Merrill, U. S. Food and Drug Inspection Station, U. S. Apprais- er’s Stores, San Fran- cisco, Cal L. Patton, U. S. Food and Drug Inspection Station, Federal Building, Buf- falo, N. Y. H. W. Haynes, U.S. Food and Drug Inspection Station, Broad Exchange Building, Boston, Mass. H. B. Mead, U. S. Food and Drug Inspection Station, U. S. Apprais- ee Stores, Philadelphia, ‘a. Ee AC Mepper=. 22-252 = T. G. Gleason, U. S. Food and Drug Inspection Station, U. S. Apprais- er’s Stores, New York, NERY: MALIC ACID per cent 0.29 0.29 0.32 CITRIC ACID percit 0.3 0.33 0.36 408 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 3 TaBLb 2.—Concluded. SaMPLeE No. 3. Sampce No. 4. (Present: Malic acid, 0.37 per cent; (Present: Malic acid, 0.35 per cent; citric acid, 0.29 per cent.) citric acid, 0.27 per cent.) ANALYST we bast ANALYST a ae per cent| per cent per cent| per cent Raymond Hertwig________- 294.022 718. OD iferrsll. -.- 2222 eee 0.30 | 0.21 0.29 | 0.24 0.32 | 0.21 P. EE Gowen. 220 Se G26") O238a Mia Pattons=-- --- aso seeee 0.29 | 0.23 0.26 | 0.24 | GER Ol & Rains Recetas LS ee 0:35") (0:23) SEC W,.baynes_ 22-2 ee 0.22 | 0.20 =e | 0:26 0.24 | 0.22 soon | (0124 EE Aen per pees oe 03323) (0'240 1 BEB. Mead: ==. =) 22 eee 0.27 | 0.19 0.32 | 0.19 O26tie ee 0:29) == M:. 'BaRongh: 224% Bie he =2=soiy O95) HAs Reppere= =) =. aaa 0.31 | 0.24 eee |) O20 = G: Weitramors eure ee nn O32 O' 23h G. Gleasone os sesso 02g ree 0.33 | 0.22 0.28 | 0.21 0.27 | 0.21 Four sets of jelly samples were accordingly prepared, with the assist- ance of Mr. M. N. Straughn, using acid-free orange pectin made in the laboratory as a base. The finished products were colored slightly with caramel and contained between 50 and 55 per cent of total sugars and the amounts of acids shown in Table 1. These samples were sent to the collaborators with the request that malic acid be determined in Samples land 2 by Method IJ, and in Samples 3 and 4 by Method III, and that citric acid be determined in all samples. The results reported are collected in Table 2. It was pointed out in last year’s report that the solubility of barium citrate in alcohol will tend to lower the percentage recovery in the case of the citric acid determination. The solubility of barium malate will produce a similar effect in the case of Malic Acid Methods IT and II. Furthermore, the effects of large percentages of sugar are likely to be such that a complete precipitation of the acids in jelly samples can not be expected. A series of determinations of citric acid in solutions con- taining amounts of sugar varying from 0.5 to 20.0 per cent were reported last year and showed a slight but undoubted decrease in percentage recovery, increasing with the percentage of sugar present. In the case of the samples containing tartaric acid also, the recovery of malic cistietitetheeensieteiemmenniioe a
  • SECOND DAY. TUESDAY—AFTERNOON SESSION. REPORT ON SPICES. By H. E. Srxpatu! (Weikel & Smith Spice Co., Philadelphia, Pa.), Associate Referee. The work has been limited to a continuation of the study of the associate referee’s modification of the distillation method for water in whole spices. Samples of Zanzibar cloves, Jamaica allspice and Lam- pong pepper were sent to fifteen collaborators with a copy of the method, and a photograph of the apparatus as used by the associate referee. Comments and recommendations were requested. Moisture in spices. ANALYST Beane Eliott, Food and Drug Department, Vermilion, . Dak. T. G. Gleason, U. 8S. Food and Drug Inspection Station, Transportation Building, Chicago, Ill. J. H. Bornmann, U. 8S. Food and Drug Inspection Sta- tion, Transportation Building, Chicago, Ill. W. C. Geagley, Food and Drug Department, Lansing, Mich. M. B. Porch, H. J. Heinz Company, Pittsburgh, Pa. R. J. Quinn, Morris and Company, Union Stock Yards, Chicago, Ill. C. H. LaWall, 39 South Tenth Street, Philadel- phia, Pa. W. B. Smith, Bureau of Animal Industry, Kansas City. Kans. C. L. Black, U. S. Food and Drug Inspection Station, U.S. Appraiser’s Stores, Philadelphia, Pa. G. N. Watson, University of Kansas, Lawrence, Kans. ZANZIBAR JAMAICA | LAMPONG CLOVES ALLSPICE | PEPPER per cent percent | per cent pal: 6.20 med usr 6.10 6.50 ae 6.58 6.80 12.4 7.40 9.40 12.2 7.80 | 10.40" 13.2» 7.50 | 13.00 7.90 5.44 6.60 8.00 5.20 7.40 11.60 7.20 8.80 12.20 7.20 8.60 wa 7.00 7.20 Bia 7.00 | 7.20 12.90 7.65 9.90 12.90 7.65 | 9.90 14.00 | 8.50 | 10.50 13.701 } 8.50 9.20 13.10 8.60 9.60 13.30 8.30 | 10.70 13.90 8.20 | 10.40 11.00 9.00 9.85 10.80 9.10 | 10.00 * Followed directions without drying connections. e Heated 45 minutes 170—180°C. © Continued distillation for 1 hour. 4 Distilled 45 minutes. 1 Present address, Austin, Nichols & Co., Brooklyn, N. Y. 427 428 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 The method sent out follows: Place 50 grams of whole spice in a distillation flask with 150 cc. of kerosene, whirl the flask several times to bring the oil into contact with each particle of spice. Place the flask on an asbestos board cut so that the bottom of the flask extends below the surface. Place a wire gauze with an asbestos center about 4 inch below the bottom of the flask. The object is to keep the flame out of direct contact with the flask. The asbestos board serves to keep the heat uniform. Connect the flask directly with a vertical condenser. Insert a thermometer through the stopper of the distillation flask extend- ing into the oil. Adjust the flame so that about 20 minutes will be required to reach the temperature of 170°C. and collect the distillate in a graduated cylinder or burette. Extinguish the flame, after which the thermometer will show a slight gradual increase in temperature. As soon as the water stops dropping from the condenser tube, which usually requires from 4-6 minutes, the operation is complete. Multiply the yolume of the water layer by 2 to obtain the percentage of moisture. DISCUSSION. The difference in the results reported by the different analysts may be in part explained by the difference in the apparatus used, and the lack of experience with the method. Consideration should also be given to the fact that all the samples were not examined by the analysts for some time after receipt. It is possible therefore that some moisture was lost previous to analysis. It seems that the method warrants further study, particularly as to size and dimensions of the apparatus and the length of time of heating. The associate referee’s attention was called to the necessity of having a definite method of sampling spices and preparation for analysis. This is a most important proposition, since some spices have a tendency to separate after being ground. RECOMMENDATIONS. It is recommended— (1) That the associate referee’s modification of the distillation method for water in whole spices be given further study with particular reference to the size and dimensions of the apparatus and length of time of heating. (2) That the subject of sampling and grinding and the preparation for analysis of each spice be studied. 1920| PATTEN: REPORT ON BAKING POWDER 429 REPORT ON BAKING POWDER. By H. E. Patren (Bureau of Chemistry, Washington, D. C.), Associate Referee. Following the recommendations of the referee for 1915, a further study of the value of the Exner and Wichmann methods for lead determina- tion in baking powders has been conducted. Last year these methods were studied for their value in determining lead quantitatively from comparatively simple solutions. This year the study has been con- tinued, using synthetic samples of phosphate baking powders where, in addition to the quantitative determination of the lead itself, the further difficulties of separating the lead from the complex mixture of ingredients have been encountered. DESCRIPTION OF SAMPLES. The samples sent out consisted of a synthetic baking powder mixed from ingredients in which the lead content had been previously de- termined. Each sample weighed 100 grams and consisted of 56 grams of monocalcium phosphate, 25.5 grams of sodium bicarbonate and 18.5 grams of starch. The ingredients were carefully mixed, and the collab- orators instructed to use the entire contents of an individual sample can for each determination. The lead content of the baking powder, calculated from the determinations on the individual ingredients as made by Mr. Seeker and Mr. Chittick, was 7 parts per million, or 0.0007 gram of lead in each 100 gram sample. DESCRIPTION OF METHODS. The methods used in the collaboration were the Exner method (modi- fied for a 100 gram sample) and the Wichmann method'. Method II was used as published. Method III was modified, using a 100 gram sample and one-half the quantity of all reagents. REPORTS OF COLLABORATORS. Reports were received from five collaborators. A summary of the results of the lead determinations is given in Table 1, as parts per million and as actual weight of lead in mg. per 100 grams of sample. From an examination of Table 1, it will be seen that the results vary greatly, not only as a whole, but in duplicate samples examined by the same analyst. The variations between determinations are as pronounced in one method as in the other. The results with the Exner method run about twice as high as those with the Wichmann method and about twice the actual lead content. The average of the results with the 1 Assoc. Official Agr. Chemists, Methods, 1916, 348. 430 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 Wichmann method agrees fairly well with the actual content of lead in the sample, but due to the wide variations in individual results the average can not be considered of much import. TABLE 1. Lead in phosphate baking powder Sample No. 161, as reported by analysts*. EXNER METHOD WICHMANN METHOD ae Mg. of P Mg. of P jsad per | Paritgn’ | teed per | Patton” J, O. Clarke, Department of Agricul- 2.0 20 1.4 14 ture, Atlanta, Ga. 3.1 31 ae 8 ie bet 6 6 oan os 0.4 4 aes ity 0.3 3 L. F. Hoyt, Larkin Company, Buffalo, 0.0 0 0.0 0 INDY 0.0 0 0.0 0 J. R. Davies, Calumet Baking Powder 1.2 12 0.5 5 Co., Chicago, Il. 1.4 14 1.0 10 A. L. Burns, U.S. Food and Drug In- 1.8 18 0.4 4 spection Station, U. S. Appraiser’s 1.0 10 1.2 12 Stores, New York, N. Y. E. R. Lyman», U. 8S. Food and Drug 0.0 0 0.0 0 Inspection Station, Arcade Annex 0.0 0 0.0 0 Building, Seattle, Wash. « Lead content of sample No. 161 as determined from analysis of separate ingredients was 0.7 mg. lead per 100 grams or 7 parts per million. - : Each 100 grams of sample contained 56 grams of calcium monophosphate, 25.5 grams of sodium bi- carbonate, 18.5 grams of cornstarch. » Since deceased. The collaborators were also asked to report on the time consumed in operations and the amount of wash water used in the various steps. The time consumed varied greatly with the different analysts, and of course depended upon the amount of other work carried on at the same time. However, a comparison of the relative time consumed for the Exner and for the Wichmann method for each analyst shows that in every case the time necessary with the Exner method was from two to three times that required by the Wichmann method. The reports on the volume of water used in washing precipitates showed excessively large quantities needed with the Exner method. The average amount of water in this method was 8 or 9 liters, most of which was used with the sulphid precipitates. Since the solubility of lead sulphid is 0.15 mg. per liter of solution saturated with hydrogen sulphid?, it can readily be seen that there is bound to be a loss of lead 1 George von Hevesy and Friz Paneth. Z. anorg. Chem., 1913, 82: 323. 1920) PATTEN: REPORT ON BAKING POWDER 431 which will be a large proportion of the total lead present, especially where such small quantities as 20 parts per million are involved. DISCUSSION OF THE METHODS AND DIFFICULTIES ENCOUNTERED IN THEIR USE. Exner method.—The collaborators found the Exner method to be very tedious and laborious. The precipitates were bulky and the quantities of wash water needed exceedingly large, thus giving opportunities for loss of lead. The iron present in the baking powder interfered greatly with the determination, contributing to the bulkiness of precipitates, and undoubtedly holding back part of the lead. In some cases the amount of lead found was low, due to being held from solution by the iron; in others the high lead results were due probably to the presence of iron salts with the lead chromate weighed. The opinion of every collaborator is that, because of the excess of time, labor and materials required, the Exner method is impracticable. Last year this method was tentatively recommended because it showed its reliability in determining lead quantitatively in fairly simple solu- tions. The results of the work this year show that, when the complica- tions due to the complex mixture of ingredients in baking powder are involved, the method is not accurate. With these considerations it would seem useless to carry on further work with the Exner method. It is recommended that the Exner method be dropped from the methods of the association and no further study be made of it. Wichmann method.—In the work with this method, bulky sulphid precipitates were also encountered. The bulkiness is caused largely by a precipitation of phosphates when the solution is made slightly alkaline, as called for by the directions given in the method. Several of the col- laborators suggest that the sulphid precipitation be carried out with the solution slightly acid, in order to avoid this difficulty. The large amount of iron present gave difficulty, and led to inaccurate results. Mr. Chit- tick and Mr. Seeker both suggest treatment of the combined iron and lead sulphates with dilute sulphuric acid, and the separation of the lead sulphate by the addition of alcohol. While this procedure would prob- ably eliminate the iron, it adds to the time required, which is a disad- vantage. The general opinion of the collaborators is that this method is a fairly quick one, and that with a few changes it has possibilities of becoming an accurate and yaluable method. The referee recommends that a further study of the Wichmann method and modifications for its improvement be made. 432 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 NEW METHODS. The deposition of the lead in baking powder by electrolysis offers a means of shortening the determination. Dr. T. J. Bryan (Calumet Bak- ing Powder Company, Chicago, Il.) has submitted a modification of the Corper method for electrolytic determination of lead, which he has found very satisfactory when applied to phosphate baking powders. It is recommended that this method be studied during the coming year. ACKNOWLEDGMENT. The associate referee wishes to thank the collaborators and analysts for their interest and for the time generously given to this collaboration, and Messrs. A. F. Seeker! (U. S. Food and Drug Inspection Station, U. S. Appraiser’s Stores, New York, N. Y.), J. R. Chittick (Jaques Manufacturing Company, Chicago, Ill.) and H. A. Lepper (Bureau of Chemistry, Washington, D. C.) for their assistance in determining the lead content of the separate ingredients from which the collaborative samples were made up, and especially Mr. G. H. Mains (Physical Chemical Laboratory, Bureau of Chemistry, Washington, D. C.) for assistance on the referee’s report. RECOMMENDATIONS. It is recommended— (1) That the Exner method for the gravimetric determination of lead (now a tentative method) be dropped from the association methods, and that no further study of it be made. (2) That a further study be made of the Wichmann method and modifications for its improvement. (3) That a study be made of Bryan’s modification of the Corper method for the electrolytic determination of lead in baking powders. No report on meat and fish was made by the associate referee. REPORT ON FATS AND OILS. By R. H. Kerr (Bureau of Animal Industry, Washington, D. C.), Associate Referee. Two methods were studied: (1) A modification of the present pro- visional method for detection of the adulteration of lard with solid fats based on the work of Bémer?; and (2) a modification of the potassium salt-acetone method for the separation of solid and liquid fatty acids*. 1 Since deceased. 2 Z. Nahr. Genussm., 1913, 25: 367; 26: 559. 2S. Fachinini and U. Dorta. Chem. Reo. Fell. Hanz. Ind., 1914, 19: 77. 1920} KERR: FATS AND OILS 433 The list of would-be collaborators was divided and samples for one method sent to one list and those for the other method to another. MODIFIED METHOD FOR DETECTION OF ADULTERATION OF LARD WITH SOLID FATS. This method represents an attempt to modify the present provisional method, taking into consideration Boémer’s discovery of the specific glyceride in lard which is responsible for the difference in melting points of the crystallized glycerides of lard and tallow. Bomer, it will be remembered, discovered that lard does not contain tristearin, but that its principal saturated glyceride is a-palmito-distearin, while beef and mutton tallow have as their principal saturated glycerides 8-palmito- distearin, and tristearin. This difference in the character of the saturated glycerides is the fundamental fact which lies at the bottom of all the methods ever devised for detecting the adulteration of lard with solid fats. In the light of his new knowledge, BOmer devised a method for detecting tallow in lard which, while based on sound principles and accurate chemical knowledge, was yet cumbersome and inefficient. The character of the method was such, however, that by a slight and simple modification it could be made to serve as an admirable check on the Emery method. The modified method sent out to the collaborators for study was as follows: Weigh out 5 grams of the filtered fat into a glass-stoppered cylinder graduated to 25 ec., add warm acetone until the 25 cc. mark is reached. Shake the cylinder until the contents are thoroughly mixed, and then let stand in a suitable place at 30°C. After 18 hours remove the cylinder and carefully decant the supernatant acetone solu- tion from the crystallized glycerides, which are usually found in a firm mass at the bottom of the cylinder. Add warm acetone in three portions of 5 cc. each from a small wash bottle, care being taken not to break up the deposit while washing and decanting the first two portions. Actively agitate the third portion in the cylinder, and by a quick movement transfer it with the crystals to a small filter paper. Then wash the crystals with five successive small portions of the warm acetone with the use of the wash bottle and remove the excess acetone from them by suction. Transfer the paper with its contents to a suitable place, spread it out, and break up any large lumps of the glycerides by gentle pressure. When dry thoroughly comminute the mass and deter- mine the melting point of the crystals. After the melting point of the crystallized glycerides has been determined, transfer them to a 50 cc. beaker, add 25 ce. of approximately N/2 alcoholic potassium hydroxid and heat on the steam bath until saponification is complete. Pour the solution into a separatory funnel containing 200 cc. of water, acidify, add 50 cc. of ether and shake. Draw off the acid layer and wash at least three times with water. Transfer the ether solution to a clean, dry 50 cc. beaker, drive off the ether on the steam bath and finally dry the acids at 100°C. Allow the acids to stand for at least 2 hours. Dry and determine the melting point in the same manner as described for crystals. If the melting point of the glycerides plus twice the difference between the melting point of the glycerides and the melting point of the fatty acids is less than 71°C., regard the lard as adulterated. 434 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. ITI, No. 4 The composition of the five samples sent out for collaborative work was as follows: Sample 1—Lard + 3 per cent oleo stearin. Sample 2—Lard + 3 per cent hydrogenated cottonseed oil. Sample 3—Pure lard. Sample 4—Lard + 5 per cent hydrogenated whale oil. Sample 5—Lard + 10 per cent beef tallow. Cooperative work on the determination of melting point, etc. ANALYST AND SAMPLE NUMBER out vrai WS A+2 (A—B) REMARKS W. B. Smith, Bureau of Animal ‘ A 7 Industry, Kansas City, Kans. ( & c DAMpPle i eater ee ho 61.4 57.2 69.8 Adulterated Sample? eee ee 61.2 60.7 62.2 Adulterated Samples ee tiec ste 28g 63.6 57.3 76.2 Pure Sample: 42 os wee eee ee 60.6 56.6 68.6 Adulterated Samples. 2 _ oe eee 61.7 58.2 68.7 Adulterated Cc. T. Allcutt, Bureau of Animal Industry, Kansas City, Kans. Sample Din stk ree. 1. 1 61.5-61.6 | 58.4 -58.0 | 67.7-68.8 | Adulterated DAMple eee eee eee 61.0-61.2 | 60.5 -60.7 | 62.0-62.6 | Adulterated Sample (3 B42 ie sre we 63.4-63.6 | 57.2 -57.4 | 75.8-76.0 | Pure Sample 464262. wete seh 59.5-59.8 | 56.75-56.7 | 65.0-66.0 | Adulterated DAM plone eee 61.6-61.6 | 58.6 -58.2 | 67.6-68.4 | Adulterated L. B. Burnett, Bureau of Chem- istry, Washington, D. C. Samples se sores ee ae 61.2 58.0 67.6 Adulterated Samples eee eee eee 60.7 60.2 61.7 Adulterated Nampleises. Sees ee 62.8 56.4 75.6 Pure Samplei4: #4 s3- ne oihs Lak 59.8 57.2 65.2 Adulterated j Sample ease soe eee 61.4 59.0 66.2 Adulterated | T. R. LeCompte, Bureau of Ani- ae Industry, Washington, Samples) wees i 8 ae er ae 62.2 58.8 69.0 Adulterated Samplei2n ses eee ee toe 62.0 61.3 63.4 Adulterated Sample ssee. eee ee en 64.0 58.6 74.8 Pure Sample sea & - Oni habe oi 61.0 57.8 67.4 Adulterated Sample/h see) ees Ae 62.3 58.4 70.1 Adulterated : R. H. Kerr, Bureau of Ani- i mal Industry, Washington, \ NG: } Sampled eees ae dosb os 61.8 58.6 68.2 Adulterated ( Samplei2.- sce ee tc 2 61.6 61.2 62.4 Adulterated ‘ sample do i325 eo cee 64.0 58.4 75.2 Pure MAO DIe ae oka eee 60.8 57.6 67.2 Adulterated | Sample pi 9222. aa See 62.2 58.2 70.2 Adulterated 2 All five samples were correctly reported by all collaborators. [1920] KERR: FATS AND OILS 435 POTASSIUM SALT-ACETONE METHOD FOR THE SEPARATION OF SOLID AND LIQUID FATTY ACIDs. In this work it was found necessary to change the method slightly after it had been sent out. In the revised form the method is as follows: Saponify a sufficient quantity of the oil or fat to set free the fatty acids. The process of saponification and preparation of the fatty acids used for the titer test may be employed. Take particular care to avoid overheating or scorching the soap or fatty acids. Next determine the iodin number of the mixed fatty acids. Dissolve 5 grams of the fatty acids in 150 cc. of pure, warm acetone, add slowly, drop by drop, a sufficient amount of N/2 potassium hydroxid solution to precipitate the estimated amount of saturated fatty acids present, and a small amount in excess. Calculate the amount of potassium hydrate required from the iodin number and the known or supposed iodin number of the liquid fatty acids of the fat or oil under examination. Allow the mixture to cool to room temperature, place in ice-water and finally leave the flask in ‘ ice-water overnight, or for at least 3 hours. Filter off the solution from the precipitated potassium salts and wash the latter twice by decantation and twice on the filter with cold acetone. Wash the precipitate, with hot water, into the flask in which pre- cipitation took place, add sufficient dilute sulphuric acid to decompose the salts, heat until the separated acids form a clear layer on top of the acid solution, cool, pour off the acid solution, add water, heat until the acids melt and form a clear layer, again cool, wash, transfer to a tared beaker, dry, and weigh. Pour the acetone filtrate and washings into a large separatory funnel, add 500 ce. of water and sufficient dilute sulphuric acid to break up all potassium salts, add 200- 300 cc. of ether, shake, draw off the acid layer, wash the ether layer three times with water, evaporate the ether and dry under vacuum or in a current of carbon dioxid or hydrogen and finally determine the iodin number of the liquid acids. A sample of cottonseed oil and a sample of inedible tallow were sent to each collaborator. Collaborators were requested to determine the percentage of solid acids and iodin number of the liquid acids on each sample and if possible to compare the results obtained with those obtained by the present provisional method. The following results were reported: Cooperative work on liquid and solid acids. SOLID ACIDS | IODIN NUMBER OF LIQUID ACIDS ea Tallowne} ae Tallow Seat A. G. Woodman, Massachusetts per cent per cent Institute of Technology, Bos- 58.8 | 31.9 90.0 147.7 OT FY Oa | 60.6 31.5 88.7 140.3 Pe Birnett. 2 2 | 38.34 17.24 66.24 80.00 BB Burnett 2) Soe sy 18 45,27 24.19 | 92.00 143.30 45.78 23.96 44.82 | Reenter ee 48.07 Bae |. asay 141.50 ® By tentative lead-salt-ether method. 436 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 While these results may be regarded as promising and while experience with the method appears to justify further study, it is clearly evident that the method as presented here is not capable of taking the place of the present tentative lead-salt-ether method. It is regarded as worthy of further study. RECOMMENDATIONS. It is recommended— (1) That the present provisional method for the detection of beef fat in lard be changed to permit the preparation of the glycerides by crystal- lization from acetone at 30°C. or from ether at 18 to 20°C. at the option of the analyst, instead of only from ether at 18 to 20°C. as at present, and that the second paragraph of the method studied, that dealing with the preparation and determination of the melting point of the fatty acids, be made a part of the method. (2) That the potassium-salt-acetone method for the separation of liquid and solid fatty acids be given further study. REPORT ON DAIRY PRODUCTS. By Jutrus Hortvet (State Dairy and Food Commission, St. Paul, Minn.), Associate Referee. The work of the past two years has included— (1) A further study of the Roese-Gottlieb method as applied to ice cream, milk powders, and malted milk. (2) A further study of modifications of the Babcock centrifugal method as applied to evaporated milk. A description of the following methods was forwarded to the col- laborators— (1) The Roese-Gottlieb method for evaporated milk and condensed milk. (2) The Manchester modified Babcock test. (3) The Hunziker modified Babcock test. MANCHESTER METHOD. Weigh 9 grams of the sample into an 8 or 10 per cent milk test bottle and set the bottle into a bath of ice-water until thoroughly chilled. Add 7.5 cc. of sulphuric acid (sp. gr. 1.84). The acid should be poured as rapidly as practicable into the milk bottle, which is inclined to one side in order that the acid may settle to the bottom of the sample with as little admixture as possible. The contents of the tube are then immediately mixed and the shaking continued until the mixture is homogeneous. 1920) HORTVET: DAIRY PRODUCTS 437 Allow the test bottle to stand at room temperature for 15 minutes and fill nearly to the base of the neck with hot water, thoroughly mix and submerge in a boiling water bath for 15 minutes, then centrifugalize for 7 minutes at about 1200 revolutions per minute. Add hot water to make the fat column rise into the scale portion of the neck, and centrifugalize an additional 2 minutes. Read the fat column ‘‘a” from the extreme bottom to the extreme top, and “‘b” from the extreme bottom to the lower line of the upper meniscus. Multiply the reading “‘b’’ by 2 and add 0.15 to the result to obtain the per cent of fat in the sample. Also obtain the per cent of fat by adding readings Bageand Dir. Nore.—Better results are obtained if the sulphuric acid employed is also cold when first added. If the weather be warm the acid bottle should be kept in a refrigerator, or a small supply may be chilled in a bottle by submerging in ice-water. HUNZIKER METHOD. Weigh 4.5 grams of the well-mixed sample into an 8 or 10 per cent milk test bottle, add 17.5 cc. of water, then add 17.5 cc. of sulphuric acid (sp. gr. 1.84), and shake until the curd in the test bottle is completely dissolved. Centrifugalize at a speed of about 1200 revolutions per minute for 5 minutes. Mix equal portions of water and sulphuric acid in a glass beaker, fill the test bottle to the zero mark with hot diluted acid and centrifugalize an additional 2 minutes, then add hot water to make the fat column rise into the scale portion of the tube and centrifugalize 1 minute. Read the fat column from the extreme bottom to the extreme top, and multiply the reading by 4. The following samples were sent to the collaborators: (1) Plain ice cream; (2) evaporated milk; (3) sweetened condensed milk; (4) dried milk; (5) malted milk. 438 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 4 TABLE Results by Roese- SWEETENED CONDENSED ICE CREAM MILK ANALYST Official Mojonnier Official Mojonnier method apparatus method apparatus per cenl per cent per cent per cent W. L. Adams, State Board of Health, 7.37 — SEIS, —— Concord, N. H. 7.49 ee See ss 7.37 ae pels ae V. B. Bonney, U.S. Food and Drug In- Had sepjarated 9.46 Bb spection Station, U. S. Appraiser’s 9.57 Des Stores, San Francisco, Cal. E. M. Bailey, Agricultural Experiment | Badly ch/urned sour aes Station, New Haven, Conn. ts Boren F. E. Schunk, Wisconsin Condensed bs aie ofc Bos 9.672 9.658 Milk Co., Burlington, Wis. ete ane 9.680 9.664 R. Hertwig, U. S. Food and Drug In- Fat sepajrated in ae eS spection Station, U. S. Appraiser’s lumps es aie Stores, San Francisco, Cal. Carnation Milk Products Co., Ocono- worst Jase bee: one mowoc, Wis. D. G. Morgan, Agricultural Experi- 7.58 ves 9.35 Soha ment Station, Stillwater, Okla. 7.68 co 9.35 apr? J. F. Snell, Macdonald College, Que- 7.74 srs 9.66 =e bec, Canada. 7.83 we 9.54 mee E. H. Berry, U.S. Food and Drug In- 7.68 are 9.40 se spection Station, 1625 Transporta- 7.61 Cees 9.45 aS tion Building, Chicago, Ill. cers == 9.50 a C. C. Forward, Inland Revenue De- 7.70 5 ae pa aah sacle partment, Halifax, N.S. 7.76 i. eit shoe Mat ose iS as Chil eee 2 ae ee Mojonnier Bros. Co., Chicago, Ill. 2s dealt. oe 9.687 1S a eae 9.683 David Klein, [inois Division of Foods 7.80 a 9.77 sae and Dairies, Chicago, Il. 7.82 Stee 9.77 SiS M. L. Jones, Sears, Roebuck & Co., 7.072 Condition 9.64 Jee Chicago, Hl. 7.07" unsatis- 9.64 i= 7.02" | factory 9.62 Bes C. N. Austin, Sears, Roebuck & Co., 6.87" | Condition} 9.62 aoe Chicago, Il. 6.875 unsatis- 9.59 aa 6.82" factory 9.56 =88 ® Not included in maximum and minimum results. 1920) 1 Gottlieb method. EVAPORATED MILE Official method per cent 7.78 7.85 7.98 7.678 7.64 7.90 7.95 HORTVET: DAIRY PRODUCTS DRIED MILK Mojonnier | Alkaline Acid Mojonnier apparatus | extraction | extraction apparatus | per cent per cent per cent per cent ae 1.39 1.43 = See 1.14 1.3 aoe see 1.13 1.34 om ——~ 1.45 1.39 a Spee 1.32 1.48 oak — 1.40 1.49 a2 7.888 1.083 1.362 1.098 7.892 1.088 b= =* 1.084 pth 128 | 1.50 me: wee os 1.23 1.47 aes = 7.86 1.31 = 1.31 Faas 1.45 1.36 ate as 1.47 1.29 = sede 1.15 0.90 Ss Sc 1.10 0.99 — ene 1.00 eee St2 220 1.30 1.25 me as ae 1.40 1.25 mares a 1.17 1.40 —— sacs 1.35 1.40 at aes, 1.23 1.38 ae ta 1.22 1.50 a (aoe Le 7.891 pase — MS Tes = ue. 1.31 Jane 5 Be 14 1.09 1.24 ip io ae 1.13 1.30 a ghee soe 1.27 Hous fe. 1.12 $27) | Gate ae 1.09 1.25 week tee 1.24 #3 439 MALTED MILK Alkaline Acid Mojonnier extraction extraction | apparatus per cent per cent | per cent 7.38 7.72 Been 6.98 7.84 oe 6.92 7.76 aut 7.52 7.70 anes 7.50 7.60 ate 7.58 7.70 ae 7.23 ase 7.38 7.33 elas 7.46 7.88 7.26 aS 7.82 7.30 oS 7.04 Pas 7.04 6.72 7.26 mipae 6.80 7.14 eet 7.26 6.70 Bee 6.80 6.62 a 712 = Ste S52 7.60 7.50 oe 7.70 7.50 oe 7.65 — Lane 7.26 8.10 =e 7.26 8.20 mates 7.50 8.14 a 7.26 8.20 ee 7.50 Soe ee20 7.50 ee seek 7.70 suse ee 7.70 wees ee 4 7.16 7.18 aes 7.12 7.22 Sane 7.18 7.34 uses 7.16 7.30 mk. 7.16 7.30 SEES 7.14 7.26 boat 440 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE SWEETENED CONDENSED ICE CREAM MILK ANALYST Official Mojonnier Official Mojonnier method apparatus method apparatus per cent per cent per cent per cent J. T. Keister, Bureau of Chemistry, 7.77 2 tas! 9.675 Ese Washington, D. C. 7.82 Sats 9.691 anes e223 ee 9.744 — L. W. Ferris, Bureau of Chemistry, ee eee 9.73 anes Washington, D. C. ee Lae 9.71 — W. D. Strack, Borden’s Condensed 7.69 7.69 9.74 9.72 Milk Co., New York, N. Y. 7.68 7.69 9.72 9.72 pay, D Me 9.73 Bg H. Hoffmann, State Dairy and Food 7.65 7.64 9.75 9.62 Commission, St. Paul, Minn. 7.68 pees 9.72 9.64 7.65 ieee 9.73 =e P. J. Donk, National Canners Associa- 8.10 ae 9.88 —— tion, Washington, D. C. 8.15 ed 9.87 ok 8.09 fy se 9.88 — Masimurmn:¢ i 3 ene eee ee eee 8.15 ae. 9.88 Jee Minimum 25 Oe ee a sa 7.37 ite Sa 9.35 Sete! TABLE 2. Results on evaporated milk by modified Babcock methods. OFFICIAL METHOD (RESULTS MANCHESTER HUNZIKER AVERAGED FROM TABLE 1) ANALYST Readings Fat Reading Fat Fat averages a b a+b | bx2+0.15 a ax4 per cent per cent per cent per cent Weil. Adams2s2225 ee Las 9.50" 9.37% 324 8.208 i% oe ont 9.508 9.378 cous 9.008 a [ sete _.. | 10.00 9.898 ees 9.00" 7.87 ee ae =e a pad Be 9.008 aod V. B. Bonney- -- --- 3.80 3.60 7.40 7.35 1.90 7.60 wae 3.80 3.60 7.40 7.35 1.90 7.60 7.655 E. M. Bailey ---_-- 4.09 3.80 7.89 7.75 2.00 8.00 aa ¥ nfecs a = aoe 2.00 8.00 7.93 * Directions for method not followed; results unsatisfactory. 1920) HORTVET: DAIRY PRODUCTS 441 1.—Concluded. EVAPORATED MILK DRIED MILK MALTED MILK Official | Mojonnier| Alkaline Acid Mojonnier Alkaline Acid Mojonnier method | apparatus | extraction | extraction | apparatus extraction extraction apparatus per cent per cent per cent per cent per cent per cent per cent per cent 7.963 —— 1.225 1.43 pe 7.14 7.44 aoe 7.986 os i 1.23 1.434 eae 7.27 7.509 as 3388 pees TENG ys a aoe ee wsee Rete 7.91 aie 1.31 1.37 ea 7.65 7.97 ae 7.94 ors 1.24 1.48 woes 7.79 7.69 as ae pipet Ee 1.34 Lf oe 7.52 7.48 owas 7.92 7.91 1.30 1.40 = 7.35 7.26 was2 7.91 7.92 1.25 1.43 se 7.38 7.23 ae 3 =P ee 7.91 1.25 1.49 —a2e 7.35 7.23 seer os oes 1.28 = == 7.31 7.22 26Le -—— es 1.40 Seeu a-28 ee ae suet 7.91 7.90 1.34 1.36 —— ee anaes a 7.95 7.89 os 1.33 pat pees = a 7.95 ee cess = tel “aes is —_ eel 7.89 ves 1.44 — tes 7.46 pee ees 7.89 ee 1.41 vans =ea= 7.43 Sent Bere 7.90 2B 1.49 Bhs Se 7.45 ee 2353 7.99 Lae BS ee ae sae 22 oe a 7.78 eee 5 ate are pts ete ae vee Tasie 2.—Continued. OFFICIAL METHOD (RESULTS MANCHESTER HUNZIKER AVERAGED. FROM ANALYST TABLE 1) Readings Fat Reading } Fat Fat a b a+b bx 2+0.15 a ax4 aVECaEeS a per cent iy per cent per cent per cent F. E. Schunk_____- 3.965 | 3.76 | 7.72 7.67 O57 7.83 7.877 metert wig): = =-- - se =1e8 8.01 8.01 We 7.74 os ee bc Sf 8.04 8.00 = 7.90 7.95 aia as 7.91 7.82 sass 7.96 eee pas! aae* 8.15 8.11 ee 7.86 oe Carnation Milk Products Go:< 3 -|) =. are 8.00 aes — ae 7.86 D. G. Morgan_____ eas tale 8.50 8.50 ae BoE 7.915 ee Monell. 2 4.04 3.85 7.89 7.85 unsatisfactory 7.95 Bs) Berry, ==—--- 4.00 3.80 7.80 Certs) 1.91 7.64 7.92 > Average of several determinations. 442 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TaBLe 2.—Concluded. OFFICIAL METHOD se Set pet. By (RESULTS MANCHESTER HUNZIKER AVERAGED = FROM ANALYST TABLE 1) Readings | Fat Reading Fat Fat a b a+b |bx2+0.15 a ax4 BN) per cent per cent per cent per cent GC. GC. Forward_____ 4.00 3.74 7.74 7.63 2.00 8.00 eae 4.03 3.80 7.83 (ers) 1.94 7.76 pee 4.01 3.78 7.79 al 1.92 7.68 oe 4.00 3.74 7.74 7.63 1.92 7.68 eal 4.02 3.80 7.82 7.75 2.00 8.00 ee 4.02 3.80 7.82 645 1.94 7.76 7.97 David Klein______- 4.00 3.80 7.80 7.75 1.90 7.60 7.96 M! i. Jones=222 4.00 3.85 7.85 7.85 Pree pty En me 4.00 3.85 7.85 7.85 eae ind ee 7.85 CAN: Austin 22-32 4.03 3.88 7.91 7.91 ees pare Be ae 4.04 3.90 7.94 7.95 oe ogee 7.89 JoDaKeisten os feo ate 7.88 7.83 ene 7.98 ae ee eee 7.86 7.81 pat > 7.98 7.974 EW. Ferris) 3 = - 4.00 3.80 7.80 7.75 1.97 7.88 aoe 4.00 3.81 7.81 7.77 1.99 7.96 7.93 W.. D. Strack___-—— ee ee Ag 5 Rese 1.90 7.60 ee = eee ee ae 1.95 7.80 7.91 H. Hoffmann. _____ 4.00 3.80 7.80 7.75 1.85 7.40 7.94 4.00 3.80 7.80 7.75 1.90 7.60 ee DISCUSSION. A number of collaborators reported results by means of the Mojonnier tester. The results submitted by the collaborators are very favorable this season, also for the Roese-Gottlieb method as applied to unsweetened evaporated milk. The results on the sample of sweetened condensed milk are also quite satisfactory, although not quite so close in agree- ment as they were in connection with the work of 1915. If, however, the results submitted by two or three inexperienced analysts are eliminated the showing will be very satisfactory. A similar statement may be made with reference to the results submitted on ice cream. In fact, there is considerable encouragement in connection with the appli- cation of the Roese-Gottlieb method to plain ice cream. The results on dried milk and malted milk, of course, are not at all conclusive. There is good reason for a further continuance of a comparative study of the official Roese-Gottlieb method and the acid extraction method. It is believed that the difficulties inherent in this portion of the work 1920] HORTVET: DAIRY PRODUCTS 443 can be cleared up next year. At least, there is ample justification as shown by the reports of the collaborators for a continuance of this work. The results on unsweetened evaporated milk by methods of Manchester and Hunziker seem to favor the adoption of the Manchester method. It is not believed, however, that any modified Babcock test can be relied upon as a strictly accurate method for determining fat in any dairy product except whole milk, and possibly cream. Nevertheless, it is deemed desirable that a centrifugal method should be adopted as a provisional method with the understanding that the method is suitable under certain conditions for a rapid sorting-out test. There is a call for such a method in routine work in food laboratories, as well as in connection with manufacturing establishments. Results obtained by the Manchester method in the hands of experienced collaborators check quite nicely with the results obtained by the official Roese-Gottlieb method. Allowances, of course, must be made for the fact that measure- ments on Babcock test bottles can hardly be made closer than 0.1 per cent. Some of the collaborators report results to 0.01 per cent. but these results are obtained by averaging a number of determinations. The Babcock test has been employed during the past quarter of a century as a satisfactory, convenient and rapid method for the deter- mination of fat in raw milk. With evaporated milk products, however, considerable difficulty has been found in obtaining comparable results by this method. Doubtless, largely on account of the changes in the proteins, as a result of the heat of processing, the solution of the sample is rendered more difficult and the complete separation of the fat pre- vented. When used with products which have not been sterilized, the results may be somewhat better, but still only approximate. Furthermore, it has long been a recognized fact that the fat column in the Babcock test contains a notable percentage of impurities, which, however, are claimed to be partly compensated for by the incomplete separation of the fat from the sample. A considerable number of the modifications of the Babcock test which have been proposed in recent years direct the use of ingredients other than sulphuric acid, thus complicating the manipulation and introducing new factors into the results. Some of these modifications require the use of amyl alcohol, others employ acetic acid, and one procedure directs the addition of a small amount of glycerol. It is obvious from a consideration of the properties of these additional reagents that the fat column is very certain to be more or less contaminated, and the measurements thereby obtained entirely too high. These facts were plainly enough illustrated by the results and comments submitted by the collaborators in 1915, and there is no disputing the conclusion, therefore, that, generally speak- ing, these various modifications of the Babcock test can not be sufficiently 444 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 accurate to be relied upon for the purpose of determining the per cent of fat in an evaporated milk. This statement may be rendered more emphatic by allowing no exceptions even in the case of those methods which yield a clear separation of fat and a well-defined meniscus. For the reasons above stated, it has therefore been definitely concluded to exclude these methods from further consideration. All attempts to correct for possible errors by the application of factors or to speculate more or less about compensating conditions, are quite too trivial to be entertained in connection with investigations which are supposedly based on sound principles. The two centrifugal methods which were tried out during the past season are not subject to the criticisms which have been pointed out. Nevertheless, there is a serious doubt in the minds of many of the collaborators as to whether such methods should be admitted at all, except as rapid sorting-out tests, which, of course, are frequently called for in connection with routine work. It seems advisable to recommend the adoption of a simple modification of the Babcock test to be applied to evaporated milk, with the qualification expressed in some manner that the test is designed to yield approximately reliable results under proper conditions. A comparison of the results seems to incline to a conclusion favorable to the Manchester method. In the method of Hunziker the use of a small amount of sample neces- sitates multiplying the fat column by a rather large factor which, at the same time, results in a corresponding multiplication of possible error. This point is a serious criticism of any analytical method. Also, gener- ally speaking, the conditions incident to the application of the two methods are quite opposite in character. In the one case the proportion of acid to sample is comparatively small, while in the other the propor- tion is large; and a similar comparison may be noted with reference to the directions for mixing the sample and acid. After careful examination of all these details, the conclusion would seem to favor the method pro- posed by Manchester. The Roese-Gottlieb method when applied to evaporated and con- densed milk products is unquestionably accurate and is fairly satis- factory in respect to the time required for the completion of a determina- tion. The time element involved in this method has been very greatly reduced by the mechanical device recently put upon the market by Mojonnier Bros. Co. of Chicago, Il., and, as already stated, the method of manipulating the Mojonnier tester does not introduce any variations which are essentially different from the details of the official Roese- Gottlieb method. In our collaborative work no attempts have been made heretofore with the application of the Roese-Gottlieb method or its modifications to samples of dried milk or malted milk. A number of analysts have 1920) HORTVET: DAIRY PRODUCTS 445 apparently acted upon the assumption that a procedure which has proved satisfactory when applied to evaporated milk products might also present no serious difficulties if carried out on products which have been reduced to a very low percentage of moisture. Various modifications of the method for the determination of fat in dried milks have for some time been under investigation by chemists, both in Europe and in America, and a number of serious problems have arisen in connection with these investigations. It has been pointed out that dried milks are likely to develop sensible quantities of free fatty acids, even when the samples are not noticeably rancid, with the result that extraction from an alkaline medium is liable to yield low results for fat. These observa- tions were verified and pointed out in the following statement communi- cated by G. E. Patrick in 1915: The difference between results by acid and alkaline extraction is quite variable— sometimes nil, sometimes a sensible amount—and I therefore felt it best to advise the use of an acid extraction method in place of the regular Roese-Gottlieb procedure in all cases of analysis of dried milks. Naturally the same principle applies to malted milks, but to a much less degree. The Eccher! method, which is only a slight modification of the Ratz- laff? method for cheese, has some advantages wherever much sugar is present and, doubtless owing to the lower temperature of heating with the acid, results in a cleaner separation of fat. Siegfeld* first made men- tion of a difficulty in obtaining full results for fat in old samples of dried milk, and Eccher has pointed out that, while it is impossible by the Roese-Gottlieb method to obtain good results on such samples (because of the presence of free fatty acids), correct results can be obtained by an acid extraction method which is only a slight modification of the Bondzynski‘ method for cheese. The only essential change consists in heating the sample with hydrochloric acid to 80°C. instead of to boil- ing temperature. This results in less carbonization and a consequently cleaner and more complete extraction. The working details of Eccher’s method may be stated briefly as follows: Heat 1 gram of the dried milk with 10 cc. of hydrochloric acid (sp. gr. 1.125), either in a small-lipped beaker with a stirring rod or in the Rohrig tube, by means of a water bath at 80°C., for 15-25 minutes (being certain that the curd is well dissolved). Cool, add 10 cc. of alcohol, using the alcohol and the ethers in transferring to the Rohrig tube, if the heating is done in a beaker. After mixing the alcohol, add 25 cc. of ethyl ether, shake well, then follow with 25 cc. of petroleum ether, and proceed as in the Roese- Gottlieb method, making the two subsequent extractions with 15 cc. of each of the ethers as usual. 1 Arch. Chem. Mikros., 1913, 6: 305. 2 Z. Nahr. Genussm., 1904, 7: 409. 3 Milchwirlschaf!. Zentr., 1910, 6: 352. * Chem. Weekblad, 1904 1: 424. 446 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. II], No. 4 A modification of this method was later proposed by C. H. Biesterfeld, formerly of the Dairy Laboratory, Bureau of Chemistry. The details of the method may be described as follows: To 1 gram of dried milk in a small beaker add 9 cc. of water and 2 cc. of ammonia, and stir with a rod until all lumps are disintegrated. Warm slightly to aid the solu- tion, transfer to the extraction tube, add 10 cc. of alcohol and mix. Extract with 25 ec. portions of ethyl and petroleum ethers as in the Roese-Gottlieb method. Pass the ethers through quick-acting filters into a tared flask. LEvaporate and weigh in the customary manner. Acidify the remaining extracted liquid in the Rohrig tube with 3.5 cc. of glacial acetic acid and place the tube up to the spigot in a water bath at 80°C. for about 10 minutes. By placing 2 or 3 glass beads in the bottom of the tube a quiet boiling action will be secured. Cool the tube in running water and add the alcohol to bring the volume to about the upper line of the spigot. Extract with 15 cc. each of ethyl and petroleum ether, and run off the extract in an unweighed flask. Repeat the extraction in the same manner, run off into the unweighed flask, and evaporate to dryness on the steam bath. Dissolve the small residue of fat in 10 cc. of petroleum ether. Transfer through a filter to the tared flask and wash out twice with petroleum ether. Finally, dry the total extracted fat on a warm plate, then in an oven at 120°C., and weigh. RECOMMENDATIONS. It is recommended— (1) That there be a further study of modifications of the Roese- Gottlieb method applied to plain ice cream, dried milk and malted milk. (2) That the Schmidt-Bondzynski modified method for the determi- nation of fat in cheese be subjected to study. (3) That the Manchester modified Babcock test be adopted as a pro- visional method to be applied to unsweetened evaporated milk. REPORT ON CEREAL PRODUCTS. By J. A. Le Cxierc! (Bureau of Chemistry, Washington, D. C.), Associate Referee. The following work was conducted: Moisture-—Comparison of the official method with the vacuum method, using as drying agents: (b*) Sulphuric acid; (c) calcium chlorid; (d) calcium oxid. Gluten.—(1) Comparison of the effect of washing the gluten until starch-free with a washing 1 minute shorter; (2) comparison of the tentative method (drying the wet gluten at 110°C. for 24 hours) with the Olson method (first heating the wet gluten at 180°C. for 15 minutes or until it springs, and then at 110°C. for 4-6 hours). Acidity—Comparison of the treatment of flour with water at 40°C. for 2 hours, as follows: (b2) Treatment with water at 40°C. for 1 hour; (c) treatment with water at 40°C. and letting mixture stand at ordinary temperature for 1 hour; (d) treatment with water at ordinary temperature for 2 hours. 1 Present address, Miner- Hillard Milling Co., Wilkes-Barre, Pa. 2 Letters ‘‘(b)"’, ‘*(c)"’ and **(d)"’ in the text refer to corresponding columns in the tables. 1920 LE CLERC: REPORT ON CEREAL PRODUCTS 447 In addition to these determinations a set of optional methods or tests was suggested and cooperation asked thereon. These included a study of the methods for the following determinations: Ash; phosphoric acid; soluble carbohydrates; cold water extract; chlorin (qualitative and quantitative). Furthermore, collaboration on two methods for making baking tests was requested. Two samples of flour (one bleached and one unbleached) were sent to each of the 14 chemists who were expected to assist in this work. Of the reports received from 10 collaborators 8 give results for moisture, acidity, gluten and ash; 6 for phosphoric acid; 5 for soluble carbohydrates and cold water extract; 4 for chlorid and baking. TABLE 1. Determination of moisture. (a) (b) (c) (d) ANALYST OFFICIAL METHOD? VACUUM METHOD VACUUM METHOD VACUUM METHOD (SULPHURIC ACID) |(CALCIUM CHLORID)| (CALCIUM OXID) J per cent time per cent lime | per cent time per ceni time J. H. Bornmann, | 11.95 5} hours| 11.56 1lday | 10.42 1 day 9.60 1 day U. S. Food and | 12.24 8 hours} 11.74 2days| 10.85 2days| 9.77 2 days Drug Inspection 11.89 3days | 10.78 3days| 9.93 3 days Station, Trans- 12.11 4 days | 10.96 4days| 9.66 4 days portation Build- 12.07 S5days | 10.97 5 days } 10.56 5 days ing, Chicago, Ill. 11.56 5 days F. GC. Atkinson, |} 10.60 11.90 11.63 13.15 (dried in American Hom- hydrogen for iny Co., Indian- 6 hours) apolis, Ind. R. A. Thuma, Uni- | 12.63 12.60 12.02 12.78 versity Farm, St. Paul, Minn. C. Kennedy, Uni-} 12.20 Shours| 12.45 4 days | 11.85 5 days | 13.22 6 days versity Farm, St.| 12.15 Shours| 13.90 4 days | 11.77 5 days | 13.27 6 days Paul, Minn. W. B. Smith, Bu- |} 10.00 10.24 9.02 reau of Animal | 10.65 10.12 9.02 Industry, Kansas | 10.23 9.03 City, Kans. 10.53 9.04 C. R. Smith, Food | 13.09 13.13 11.00 and Drug Inspec- tion Station, U. S. Appraiser’s Stores, New York, INS Y': L. H. Bailey, Bu-| 12.21 12.17 1 day 9.12 5 days} 12.09 1 day reau of Chem- 12.46 2 days 12.70 5 days istry, Washing- 12.88 5 days ton, D. C. L. Dunton, Agricul- 10.82 5days| 8.61 38days | 10.19 3 days tural Experiment Station, Manhat- tan, Kans. * Assoc. Official Agr. Chemists, Methods, 1916, 79. 448 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 In sending out these samples no attempt was made to transport them in air-tight containers because our object was not to see if the different collaborators would find the same amount of moisture in the flour, but to see what results would be obtained by the use of the different methods in the hands of each collaborator. MOISTURE. (See Table 1.) In five cases Method (a) gave results agreeing with Method (b). In two cases the results were too widely different. In every case Method (c) gave lower results than Method (b). Wherever Method (d) is compared with Method (b), the results show an encouraging agreement. All the resulls given in Tables 2 to 5 are on the water-free basis. TABLE 2. Determination of gluten. (On water free basis.) WET DRY (a) (b) (c) (a) (b) (c) Dried at ANAURER Gluten Dried to | Dried to | 180°C. for Tentative | washed | Tentative || constant | constant | 19 minutes methods | 1 minute | method* || weight at | weight at and less than 105-110°C.»|105-110°C.»| to constant (a) weight at 105-110°C.> per cent per cent per cent per cent per cent per cent Tee ean 40.85 | 41.45 | 41.10 || 1284 | 13.06 | 12.93 BOG] Atkmsone=- 22-5 28.4¢ p2Wisse) ||) ees 9.56° 9.758) 4teeee Oe ee ee 364 | 367 | 366 12.72 | 12.82 | 11.67 Coiceneiyeeh eee eas 38.30 | 38.56 | 3814 || 11.78 | 1218 | 1229 WE jcriih ome eee 34.95 | 32.95 | 33.7 || 11.20 | 11.00 | 11.20 Caro sain. ee 36.8 | 37.7 | 37.7 14.0 | 140 | 12.9 H. L. Wessling, Bureau of Chemistry, Wash- I LONG aes 32.5 34.2 33.3 11.6 11.9 11.8 LL; Dinton 36.7 36.6 36.7 12:25 11.94 11.63 Average: <1 =->ee 36.6 36.9 36.8 12.3 12.4 12.0 ® Assoc. Official Agr. Chemisls, Methods, 1916, 189. + Results were obtained from corresponding column under wet gluten. © Not included in average. GLUTEN. (See Table 2.) The results of the gluten determination show a very good agreement between Methods (a), (b) and (c) in the hands of any one collaborator and this applies to both the wet and the dry gluten. Too much variation between results obtained by different col- laborators is to be noted, however, in examining the results of any one method, e. g., Method (a) for wet gluten gives results varying from 32.5-40.8 per cent; Method} (b) ‘ 1920) LE CLERC: REPORT ON CEREAL PRODUCTS 449 34.2-41.4 per cent; Method (c), 33.3-41.1 per cent. For dry gluten Method (a) varies in the hands of the different workers from 11.2-14.0 per cent; Method (b), 11.0-14.0 per cent; Method (c), 11.2-12.9 per cent. ACIDITY. (See Table 3.) No appreciable differences between the results obtained by any one collaborator are to be noted in Methods (a), (b), (c) and (d). Method (c) is the simplest, consisting of treating flour with water at 40°C. and letting the mixture stand at ordinary tempera- ture for 1 hour with occasional shaking. ASH. (See Table 4.) In eight out of nine cases the use of calcium acetate gave higher results than did the provisional method. In five of these eight cases, however, the results obtained by the use of calcium acetate are only slightly greater than those obtained by the provisional method. The results show that ashing by the provisional method is the more satis- factory and correct, provided incineration is not carried on at too high a temperature. This is evident since the results on phosphoric acid are identical when it is deter- mined upon the ash obtained by these two methods. One must assume, therefore. TABLE 3. Determination of acidity. (On water free basis.) (a) (b) (©) | (da) SAMPLE WATER AT PLUS WATER SAME AS 40°C, AT ROOM ANALYST METHOD OF (a) EXCEPT ALLOWED TEMPERATURE; COMMITTEE C, | SAMPLES KEPT TO STAND ALLOWED 19159 ar 40°C. For 1 HOUR TO STAND 1 HOUR AT ROOM 2 HOURS TEMPERATURE AT ROOM TEMPERATURE per cent per cent per cent per cent Jeb borcnmmann: 2. = 8 0.109 0.113 0.110 0.098 Be vAtkinson: 22-52-2222 2: 0.163 0.158 0.146 0.152 era ihumas: 2s es 2 0.172 0.166 0.160 0.160 rmifermed ys sees dey! eae 0.127 0.119 0.125 0.119 Me enomith ste ee eet 0.095 0.089 0.084 0.091 C02, Sin ee ee 0.106 0.107 0.104 0.106 eplbs Wessling. 220.222 2. 0.137 0.134 0.131 0.129 D. H. Grant, Bureau of Chem- istry, Washington, D. C.____- 0.133 0.125 0.120 0.116 .. Dinh eee ee ee 0.153 0.153 0.134 0.124 PMrenapee se ee 0.133 0.129 0.124 0.121 a J. Assoc. Official Agr. Chemists, 1917, 3: 87. 450 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 that the results of ashing by Method (a) are better than the results obtained by Method (b) or assume that a large portion of the non-phosphoric acid material has been volatilized. The latter assumption is not logical. On the other hand it is reasonable to assume that the higher results of Method (b) are due either to incomplete incinera- tion or to the absorption of moisture by the sample before it has been weighed. SOLUBLE CARBOHYDRATES. (See Table 5.) Five reports were received. The average results from Methods (a), (b) and (c) are almost identical, i. e., 1.33, 1.388 and 1.41 per cent. In three cases Method (a) gives the lowest results; in two cases Method (b) gives the lowest results; in two cases results with Method (a) are the highest. TABLE 4. Determination of ash and phosphoric acid. (On water free basis.) ASH PHOSPHORIC ACID (a) (b) (as) (bb) (ce) ANALYST Ash bealee ios Official Calan Official Official nla ae method methodé method method! oficial method subsequently followed per cent per cent per cent per cent per cent J. H. Bornmann_- 0.472 0.540 0.222 0.233 0.222 F. C. Atkinson___- 0.442 0.570 fds SF 2552 Be? R. A. Thuma___-- 0.504 0.538 0.252 0.252 0.252 C. Kennedy - - - --- 0.495 0.502 0.246 0.252 0.248 W. B. Smith__-_--_- 0.479 0.480 0.235 0.233 0.230 H. L. Wessling--_- 0.478 0.490 wees mer Sees L. H. Bailey.----- 0.475 0.464 0.246 = ae 0.242 D: BH: Grant-22:-- 0.467 0.464 oe Bn ee. L.. Duntons-2-2+ 22 0.466 0.534 0.229 0.239 0.229 Average-.----- 0.474 0.509 0.238 0.245 0.237 ® Results in this column were obtained by determining the phosphoric acid in the ash reported by the official method, column (a) under ash. 9 b Results in this column were obtained by determining the phosphoric acid in the ash reported by the calcium acetate method, column (b) under ash, Lee © Results in this column were obtained from the ash determined by the official method, column (a) under ash, 4 Assoc. Official Agr. Chemists, Methods, 1916, 187. © U.S. Bur. Chem. Bull. 107, rey.: 21. ! Assoc. Official Agr. Chemists, Methods, 1916, 3. 1920] LE CLERC: REPORT ON CEREAL PRODUCTS 451 COLD WATER EXTRACT. (See Table 5.) It is evident from the reports of the collaborators that enzym action is sufficient to vitiate the results. The extraction, therefore, should be carried out at a temperature not above 10°C. CHLORIN BLEACHED FLOUR. Two collaborators tested for chlorin qualitatively and reported unsatisfactory results. Four collaborators reported on the quantitative determination of chlorin, two of them obtaining 26 and 27 mg., respectively, per kilogram of flour, the other two obtaining 63 and 205 mg., respectively. TABLE 5. Determination of soluble carbohydrates and cold water eztract. (On water free basis.) SOLUBLE CARBOHYDRATES COLD WATER EXTRACT (a) (b) (c) | (a) (b) ANALYST Water at room Bryan, Given Sodium 0.5 per cent temperature and Straughn | bicarbonate | hydrochloric Ice-water saturated method= method> acid with toluene; method kept at room temperature per cent per cent per cent per cent per cent J. H. Bornmann __ 1.07 1.50 1.59 5.31 7.71 R. A. Thuma___-_- 1.42 1.11 1.24 6.63 6.59 C. Kennedy _-_---_- 1.52 1.02 1.24 5.53 7.07 Wier. Smith=. _ =~ 1.34 1.45 1.38 = ——- DH. Bailey_- ---- 1.32 1.86 1.60 5.81 7.15 i Dunton_.=--—-- See. aoe Bae 4.86 4.36 Average_____- 1.33 1.58 1.41 aes thse | iI * Assoc. Official Agr. Chemisls, 1916, 109. > U.S. Bur. Chem. Bull., 162: 121. DISCUSSION. The lack of agreement between results obtained by different col- laborators using the same methods suggests that the procedures of different analysts vary beyond the limit of error of the methods them- selves. It is probable that a demonstration of the methods by one skilled in their use would be sufficient to produce more harmonious results when study of the subjects is continued. RECOMMENDATIONS. It is reeommended— (1) That Method (b) for the determination of moisture in flour and similar cereal products be approved; that Method (d) and Method (a) receive further study; and that Method (c) be dropped. 452 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (2) That the method of drying wet gluten by heating to 180°C. for 15 to 20 minutes, or until it springs and then at 105° to 110°C. to con- stant weight (4 to 6 hours) be approved, and that work on wet and dry gluten and the washing of gluten be continued. (3) That Method (c) for acidity be approved, and that the methods for the determination of acidity in flour receive further study. (4) That Method (b) for ashing be approved, weighing to be made immediately upon cooling. (5) That Method (b) for soluble carbohydrates in flour be abandoned and that Method (c) be adopted. (6) That Method (a) for cold water extract in flour be approved. (7) That further study be made on the testing for chlorin in bleached flour. 1920] BIGELOW: REPORT ON CANNED VEGETABLES 453 REPORT ON CANNED VEGETABLES. By W. D. Bicetow (National Canners Association, 1739 H Street, Washington, D. C.), Associate Referee on Vegetables. Collaborative work on canned vegetables was confined to a study of the Howard method for the microscopic examination of tomato pulp. The details of this method have been published?. The limitations of the method with respect to magnifications employed, character of organisms counted and similar questions were not considered. Although the im- portance of such questions was appreciated, it was believed that a study of the technique of the method and the promotion of uniform procedure among analysts were of more immediate importance. In the early spring of 1916, a letter was written to all official labora- tories which were known to examine tomato pulp and ketchup, and to all manufacturers and buyers of tomato pulp who were known to main- tain laboratories, asking collaboration in the study of the Howard method. Encouraging response was received and eight samples were sent to each laboratory which indicated a willingness to take part in the collaborative work. Reports on at least some of the samples were received from fifty-two analysts representing thirty-eight laboratories. Of these laboratories. sixteen were official, charged with the enforce- ment of Federal, State and Municipal food laws; sixteen were trade laboratories, maintained by manufacturers or dealers; and seven were commercial laboratories, accustomed to examine tomato pulp for either manufacturers or dealers or both. This work was intended to be of a preliminary nature, and collabora- tion was asked with the understanding that the results of the individual analysts would not be made public. The results received are given in Tables 1,2 and3. The laboratories and analysts are indicated by number and, where more than one analyst reported from a single laboratory, the additional analysts are indicated by letter. CHARACTER OF EQUIPMENT USED. Collaborators were asked to report the character of microscopic acces- sories they employed. Apochromatic accessories were employed by Analysts 1 to 3B, inclusive, 10, 13 and 32. All other analysts employed ordinary achromatic objectives and Huyghenian eye pieces. Those who collaborated in later work, the results of which are given in Tables 4 to 8, inclusive, were equipped with apochromatic accessories. During the season a number of the collaborating analysts brought their micro- 1 Assoc. Official Agr. Chemists, Methods, 1916, 324. 454 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 scopes to the writer’s laboratory and it was found that the accessories used by some would not give the necessary definition. CONFERENCES OF ANALYSTS. When the widely discrepant results were noted, two conferences of tomato pulp analysts were arranged, at which the analysts worked both with their own instruments and with instruments equipped with proper accessories. Much progress in uniformity was made at these conferences. DISCUSSION OF TABLES 1, 2 AND 3. As stated above, there are given in Tables 1, 2 and 3 the results obtained by collaborating laboratories on eight preliminary samples. These results showed the widest possible variation. For instance, on Sample A, the mold content reported by different analysts varied from 6 to 90 per cent of the fields. The bacteria count varied from 6,000,000 to 168,000,000 per cc., while the content of yeasts and spores varied from 1 to 208 per «s cmm. In Sample 1, the mold count varied from 4 to 100 per cent of fields, the bacteria count, from 8,000,000 to 115,000,000 per cc. and the content of yeasts and spores varied from 0 to 850 per av cmm. Similar variations will be noted in the other samples given in these tables. Of the eight samples, four were submitted to a number of laboratories in duplicate, but under different numbers. The results obtained from both examinations are given. These results illustrate the fact that good duplicates do not necessarily mean correct work. In a number of cases it will be noted that analysts whose results were far from correct checked themselves very well in examining duplicate samples, though they did not recognize them as duplicates. COLLABORATIVE WORK WITH SAMPLES OF KNOWN CHARACTER. During the tomato season just passed, a series of uniform samples was prepared under the personal observation of Messrs. B. J. Howard and C. H. Stephenson of the Bureau of Chemistry, and Mr. H. M. Miller of the National Canners Association. All three men were present during the manufacture of these samples and careful observation was made of the raw material and the method of manufacture. Several samples of the pulp were taken at different stages of the concentration. From the results given in Tables 1, 2 and 3, it was obvious that nothing would be gained by having these samples examined by the entire number of collaborators. A small number of analysts was selected whose results in the preliminary work showed a reasonably good working knowledge of the method. Several analysts were also included who, it was believed 1920} BIGELOW: REPORT ON CANNED VEGETABLES 455 as a result of the preliminary work, had been able to correct their procedure. This list includes sixteen analysts located in twelve labora- tories. Of these laboratories, four (six analysts) were official, seven (nine analysts) were trade laboratories and one (one analyst), a com- mercial laboratory. The samples were examined for the purpose of securing information on the following points: (1) Agreement between analysts working with the same samples. (2) Relation of the concentration of pulp to the microscopic count. (3) Relation of the microscopic count to the amount of rotting ma- terial in the raw product and hence to the perfection of sorting of the raw product. (4) Influence of delay in manufacture on the microscopic count. (5) Influence of the finishing operation on the microscopic count. The conditions under which tomato pulp is manufactured vary so greatly that much difficulty is encountered in appraising the value of the Howard method for determining the presence of decomposing material, and in adopting a uniform and satisfactory procedure for interpreting the results. It was necessary, therefore, to examine a considerable number of samples prepared under different conditions of manufacture. AGREEMENT BETWEEN ANALYSTS WORKING WITH THE SAME SAMPLES. The results in Tables 4 to 8, inclusive, are so arranged as to give an idea of the agreement that was obtained by the sixteen analysts to whom the samples were submitted. In addition to the detailed results of the various analysts, there is given also for each sample the average of the results reported. In accordance with the custom of the associa- tion, the more extreme results have been excluded from the averages. In some cases, new samples were sent to analysts with the request that they repeat certain determinations. In such cases, the results of both determinations are given in Tables 4, 6, 7 and 8, although the results of the first examination are excluded from the averages. In the case of Analyst 6, so many results varied greatly from those obtained by other collaborators that it was thought best to exclude all of his results from the averages. It is pointed out, on page 458. that the count of molds does not appear to be changed by the concentration of the pulp, though the bacterial count appears to be roughly proportional to the concentration. The agreement between analysts in mold count, therefore, can be studied 456 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 advantageously in Table 4, while the bacterial count can be studied better in Table 5, not taking into consideration those figures that are excluded from the averages. The detailed figures for mold count in Table 4 are especially interesting. In Series 1 and 2, in which the average mold count is 29 and 18, respec- tively, results of the individual analysts rarely vary from the average by as much as 10. While an error of this magnitude is material when considered on a percentage basis, it becomes insignificant when we con- sider the purpose of the method and that a mold count of 66 per cent is permitted. The samples mentioned represent good average pulp, and the reports from the individual collaborators show that there is no probability of their mistaking a pulp of this nature for a pulp containing molds in 66 per cent of the fields. With Series 3 and 4, the case is somewhat different. The samples in Series 3 have a mold count of about 50. Two of the analysts reported results, on one sample each, as high as 66 per cent of the fields. Samples in Series 4 have a mold count of about 65. A variation of 10 either way from this figure, that is, a range of from 55 to 75, would include a great majority of the results received, yet several collaborators reported results outside of this range. For instance, Analysts 4 and 8 reported results as low as 40 and 41, while Analysts 5, 8 and 10 reported results as high as 86 and 87. The individual results obtained by all analysts from the various samples of each series agree with each other much more closely than do the results of the various analysts. For instance, Analyst 10, who obtained the highest results, had a range of from 73 to 87; Analyst 5, whose results were also high, had a range of from 71 to 86; Analyst 8, whose results were the lowest, had a range of from 40 to 54. It should be borne in mind that the samples were all submitted to the collaborators by number, and it was not possible for any collaborator to know that the samples of any series were in any way related to each other. It is therefore obvious that, with more intensive work and more frequent checking between the different collaborators, the variation in the results of different analysts would more nearly approach the varia- tion in the results of one analyst from the examination of a single series of samples. In Table 6 the variation between the mold count of the different analysts was similar in a general way to Series 3 and 4 of Table 4. The results given in this table show that, working with a limit of 66, analysts of experience in the method are likely to condemn a sample of pulp whose correct mold count is as low as 50. Just how much of this is due to inaccuracy of the method and how much to personal equation, it is impossible to say at this time. 1920) BIGELOW: REPORT ON CANNED VEGETABLES 457 Under present conditions, therefore, it is obvious that. with a limit of 66, a sample which an individual analyst finds to contain molds in more than 50 per cent of the fields is in the danger zone. It is also obvious that in taking action on a sample whose mold count is not greatly in excess of 66, it is necessary for an official analyst to examine several samples, or to make a number of independent examinations of the same sample. The results of individual collaborators on bacterial counts on Series 1 to 4, inclusive, of Tables 4 and 5, show that, when working with samples with a low bacterial count, experienced analysts should reach the same conclusion, although the percentage of variation may be extremely large. With a somewhat higher bacterial count, the percentage of variation appears to decrease. With a high bacterial count, such as results from allowing tomato pulp to stand for a number of hours in process of manufacture, it appears that the difficulty of distinguishing the organisms from other bodies of microscopic size is greatly increased. This increases, in turn, the error of the method and the variations in the results of different analysts. It is the experience of the writer’s laboratory that, as in the case of molds, much greater uniformity can be obtained in the bacterial count by a careful study of the method. At the same time, the increase in bacteria is accompanied by the disintegration of the cells and the formation of debris to such an extent that an accurate count is difficult. This difficulty can be overcome to a certain extent by dilution of the sample, but this, in turn, introduces a proportional multiplication of error. In using the method, therefore, it must be kept in mind that the present working limits were adopted with the belief that they were really excessive and with a view to making allowance for error of method, variation in sample and personal equation of analysts. Great care should be taken, therefore, in the interpretation of results which approach closely the limit mentioned. Official analysts will be wise to confirm very carefully bacterial counts which are not far above the limit of 100, and manufacturers should view with suspicion batches of pulp whose bacterial count is reported to them as over 80. When the bacterial count greatly exceeds 100, the variation between the results of two analysts or between the results on the same sample by the same analyst increases enormously. This is doubtless partly because analysts have not given much attention to the exact count of pulps that were obviously far beyond the limit. As stated elsewhere, this difficulty can probably be overcome to a certain extent by increased dilution of the samples counted. It is unfortunate that the samples examined did not include some that were high in yeasts and spores. The maximum number of yeasts 458 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 and spores permitted by the Bureau of Chemistry is 125 per “> emm. The average yeast and spore counts of the samples included in Tables 4 to 8 varied from 6 to 27. In general it may be stated that the variation of the individual collaborators in the yeast and spore count was similar to the variation in the bacteria count in Series 1 to 4, inclusive, of Table 4, and in Tables 6 and 8. It is hoped that the work will be con- tinued and a comparative study made of a series of samples with a relatively high content of yeasts and spores. RELATION OF THE CONCENTRATION OF PULP TO THE MICROSCOPIC COUNT. In Table 4 are given results obtained by the Howard method on eight series of samples of tomato pulp taken at various stages of concentra- tion. The first sample in each series represents the unconcentrated pulp just as it flowed from the cyclone. The succeeding samples in each series were taken from the same kettle as the first sample, at varying degrees of concentration, the extent of which is indicated by the per cent of solids. The influence of concentration on the microscopic count is best shown by a comparison with each other of the average figures of the samples in each series for molds, bacteria, and yeasts and spores, respectively. The mold count is strikingly uniform throughout the various stages of concentration. The bacterial count in the first four series is also uniform throughout the different stages of concentration. In the last four series, the bac- terial count increases with increased concentration. It appears alto- gether probable that the lack of increase noted in the first four series with increasing concentration may have been due to the fact that the number of organisms present was no greater than the error of the method. In Table 5, the bacterial count of the various samples in each series is calculated to pulp of 8.37 per cent solids. This percentage of solids is equivalent to a specific gravity of 1.035, which is the concentration most commonly demanded in the trade. Although there are a few striking exceptions, the uniformity in general is so great as to suggest that the bacterial count is directly proportional to the concentration of the pulp. It will be noted that, as a rule, the average result on the first sample in each series is higher than the others. This is probably due to the fact that, in the first sample of each series, the dilution which was counted was of lower concentration than in the case of the other samples. Conse- quently, there were fewer particles of all kinds in each small square of the microscopic field and these particles were identified and counted more easily and with greater accuracy. This suggested the thought that more accurate results might be obtained by diluting highly con- ~ 1920) BIGELOW: REPORT ON CANNED VEGETABLES 459 centrated pulps and pastes further than is called for in the Howard method, and subsequent experimental work in the writer’s laboratory has established the wisdom of this procedure. It is believed that to obtain the best results the samples should be diluted to such an extent that the small squares will contain less than 10 bacteria each. In case the field shows considerable debris the sample should be still further diluted. In this way the error of the determination is greatly increased by the factor with which the microscopic reading must be multiplied. This error is more than compensated, however, by increased accuracy in identifying the organisms in the field. The results obtained from the determination of yeasts and spores, as shown in Table 4, do not indicate an increase in the number of yeasts and spores with increased concentration of the pulp. It is unfortunate, however, that all of the samples taken were so low in yeasts and spores that it is entirely possible that the error of the determination would be sufficient to conceal a greater number of those bodies in the higher concentration. It is pointed out above that, in the first four series in Table 4, which were very low in bacteria, the bacterial count is practically uniform in all stages of concentration, while in the last four series, which were high in bacteria, the bacterial count is practically proportional to the concen- tration. The fact that the yeast and spore count, which is low in all samples, is practically the same for all stages of concentration, therefore, does not warrant the inference that uniformity would obtain with different stages of concentration of a sample high in yeasts and spores. RELATION OF MICROSCOPIC COUNT TO THE AMOUNT OF ROTTING MATERIAL IN THE RAW PRODUCT. It is well known that the particles of mold disclosed by the micro- scope in the examination of a sample of pulp come from several sources. Mold sometimes clings in considerable quantity to the outside of toma- toes. While by far the greater proportion of this is removed by washing the tomatoes, undoubtedly some particles remain and are found in the pulp. Again, in view of the minute particles into which the mold is broken in the manufacture of pulp, if all the appliances with which the mold comes into contact are not kept scrupulously clean there is a possi- bility of the amount of mold in the finished product being increased from that source. By far the greater part of the mold particles in tomato pulp, however, comes from rotting portions of the tomatoes from which it is manufactured, and it is believed that. if it were possible to remove every particle of rot from the raw product, the mold content of the finished pulp, made with reasonable care in washing and sorting and reasonable cleanliness of the plant, would be very low. 460 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 It is obviously impossible to remove every particle of mold from the raw product. Even if perfection were attainable in trimming the out- side of the tomatoes, it sometimes happens that a mass of mold is found in the center of the tomato with no mark on the outside to indicate its presence. It becomes a matter of importance, therefore, to determine to what extent the mold count of the finished product is an indication of efficiency in sorting and in the elimination of rot. To study this question, it was planned to visit a number of plants during the season and to take samples from typical batches of pulp, carefully inspecting the raw product from which they were made and determining the percentage of rotting material in each. Unfortunately, time did not permit this work to be done to the extent that was planned. Only nine samples were put up in this manner, and they consisted en- tirely of trimming-stock pulp; that is, pulp prepared from the peelings and cores of the tomato cannery. Since in canning tomatoes there is not far from 50 per cent of waste, the amount of trimming-stock pulp made from a ton of tomatoes is something less than one-half (probably not far from 40 per cent) of the amount that would be made from a ton of whole tomatoes. Since the rotting portions of tomatoes are almost all on the outside, practically all of the decomposing material of the tomatoes will be included in the trimming-stock pulp, so that from the same tomatoes the amount of decomposing material in trim- ming-stock pulp is probably about two and one-half times the amount in pulp made from whole tomatoes. Therefore, if the figures given in the column headed “Rot in Tomatoes’, in Table 6 be multiplied by the factor 2.5, they will probably approximate the percentage which the rotting portions would constitute of the trimming stock from which the pulp was made. The results shown in Table 6 are too meagre and are not sufficiently consistent to enable any conclusion to be drawn as to the relation between the amount of rotting material in the raw product and the mold count of trimming-stock pulps. In a general way it would appear that the mold count of samples made from material containing a larger amount of rot was higher than in samples made from material contain- ing a smaller amount of rot. It is regretted that the volume of the work is not sufficient to permit a more definite conclusion. It should be remembered that the determination of rotting material is complicated by the fact that, in order to remove all of the rotting portions of the tomatoes, it is necessary to cut out a relatively large amount of sound material. Undoubtedly, therefore, the figure given in the column headed ‘‘Per Cent Rot” is only an approximation of the amount of tomatoes that must be discarded in removing the rot. In order to determine the amount of rotting material in the raw product, 19201 BIGELOW: REPORT ON CANNED VEGETABLES 461 the tomatoes from whose trimmings each batch was made were sampled on the sorting belt after passing the sorters. To prevent the possibility of selection, the samples were taken by removing a shoyelful at a time from the belt at frequent intervals. It was assumed that the samples so taken represented the entire batch. From these samples, the rotting material was cut out as neatly as possible, the trimmed tomatoes and the rotting portions were weighed separately and the percentage of rot was calculated. The batches of pulp made from the tomatoes so sampled were sealed in cans, processed in the usual way and samples taken for analysis. INFLUENCE OF DELAY IN MANUFACTURE ON THE MICROSCOPIC COUNT. When supplied with suitable medium, it is well known that bacteria increase with remarkable rapidity. When tomato pulp is found with a high bacterial count, the reason is sought first in delay at some stage of the manufacture of the product. It is obvious that the increase in number of bacteria will depend on many conditions which are more or less susceptible of control. The variation from plant to plant and even from day to day in the same plant is so great that it could hardly be said that any individual illustration would be typical. It was thought to be of interest, however, to call attention to one sample of pulp which was allowed to stand all day in the factory for the purpose of observing the microscopic count at intervals. The results of this work are shown in Table 7. Unfortunately, the fresh sample, whose bacterial count would doubtless have been very low. was not sampled. After Sample 26 was sterilized, the remainder of the batch was allowed to stand 6 hours, when a portion of it was removed, sterilized and numbered 28, the remainder being allowed to stand 4 hours longer, when it was sterilized and numbered 36. INFLUENCE OF THE FINISHING OPERATION ON THE MICROSCOPIC COUNT OF TOMATO PULP. The screen used in an ordinary cyclone employed for straining uncon- centrated tomato pulp from the peelings and seeds permits many black specks, consisting of particles of discolored stems, calyx and black mold, to pass through with the pulp. To remove these and improve the con- sistency of the product, many manufacturers employ a finishing ma- chine, which is provided with a much finer screen than the cyclone. The influence of this screen on the microscopic count of tomato pulp has often been a matter of discussion. Some believe that the fine screen tears to pieces clumps of bacteria so that they can be seen more readily, 462 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 4 and divides pieces of mold into a number of smaller particles so that they are distributed among several microscopic fields. Some, on the other hand, believe that the finishing machine does not change the bacterial count and that, by removing the clumps of hard rot, it actually reduces the mold count. For a study of this matter, six batches of pulp which were being manufactured in six different canneries were made the subject of study and samples of each were taken both before and after finishing. The results are given in Table 8. It will be noted that the microscopic count of the unfinished samples and the finished samples is practically identical. CONCLUSIONS. The most striking lesson from the work presented here is that the majority of analysts who profess to use the Howard method are not familiar with it; the method is purely arbitrary and an analyst can not expect to obtain satisfactory results without being personally trained by one accustomed to use the method. It is necessary that the micro- scopic accessories be well adapted to the method. It is the experience of the writer’s laboratory that from two weeks to a month of consecutive work under the personal direction of an experienced man is necessary to train an analyst in the use of this method, and that constant practice is necessary to retain a working knowledge of it. Of still greater im- portance is the aptitude of the analyst. The method is not one that all can learn to use. In a method of this nature it can not be hoped that duplicate read- ings can be obtained which will have as much uniformity as is expected in analytical chemistry. Judging by the results obtained, it is important that a number of samples be examined from a given lot of pulp or that a number of inde- pendent counts be made on it before conclusion is drawn. Working with the standard of 100 million bacteria per cc. and molds in 66 per cent of fields, it would appear from the results obtained that official analysts should check their work carefully before taking action regard- ing pulp which only slightly exceeds those limits and that manufac- turers, on the other hand, should view with suspicion any pulp whose bacterial count exceeds 80 million per cc. or which shows mold in more than 50 per cent of the fields. In this connection it should be noted that the figures given in Tables 1, 2 and 3 amply demonstrate that no credence should be given to the work of analysts who have not had personal instruction from one skilled in the use of the method. 1920} BIGELOW: REPORT ON CANNED VEGETABLES 463 The yeast and spore content of the samples examined was so low in all cases that no conclusion can be drawn regarding the ability of experi- enced analysts to check each other in samples whose yeast and spore count is relatively high. In the concentration of pulps, the mold count is not increased, but the increase in the bacterial count is probably proportional in a general way to the concentration. This increase was not demonstrated in the samples of low bacterial count, but was evident in the samples high in bacteria. It seems probable, therefore, that the fact that it was not demonstrated by the samples low in bacteria was due rather to difficulty in counting the more concentrated pulps in each series than to the fact that the increase did not occur. Unfortunately, the series of tomato pulps low in bacteria were not concentrated to a heavy paste and the data secured on the concentration of pulps of this nature is therefore not complete. Delay in the operation of manufacture of tomato pulp may, in some instances at least, cause great increase in bacterial count. The opera- tion known as “‘finishing” (straining through a very fine mesh screen) does not increase the count of bacteria, mold or yeast and spores. Note.—Throughout these tables, where more than one determina- tion is reported on the same sample by one analyst, a duplicate sample was sent him, without his knowledge of its nature, for the purpose of checking the first determination. 464 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE 1. Mold count of tomato pulps. (Expressed as per cent of fields showing molds.) ANALYST arabes ere es ear) eed ae rae ora SAMPLE we 1 57 53 60 60 67 51 = 24 1A 62 61 68 55 72 44 17 30 2 57 45 82 75 79 61 10 38 3 60 52 70 63 75 56 30 27 3A 56 46 70 66 80 60 30 37 3B 54 44 76 50 42 54 == 24 3B Se 54 2 33 Ls ae aie Bi 4 60 40 60 52 64 53 16 28 4 44 56 ay — =f fh 27 5 83 48 69 60 70 52 12 38 5 = 55 76 fe =2 a2 a 40 5A 82 42 79 65 75 53 18 31 5A an 62 79 = te a ue 38 6 25 64 70 56 32 49 23 33 7 63 36 80 30 98 32 12 12 8 68 36 52 50 52 42 8 18 8 2 40 46 = =s te Re 18 9 30 49 49 62 41 69 20 30 10 62 50 66 55 65 58 15 36 10 = 49 69 ae = ae i 23 il 56 = = = oe a ae ae 12 54 20 78 28 58 52 4 32 12 Ee 24 68 ae = ae =2 32 13 60 50 64 52 38 44 14 26 13 =2 52 2 ae ie ae =e 32 14 63 66 93 66 74 76 42 36 144A 65 78 81 79 86 72 40 36 15 80 26 84 46 54 44 12 28 16 12 24 20 10 14 30 4 4 17 70 35 55 32 58 30 20 30 18 90 = ae ae = ec =e ae 18A 83 ae ae =e oi ue = =s 19 ze 48 42 46 46 38 18 14 19A = 36 44 36 38 38 26 28 19B Ae 46 56 30. | = 34 44 16 20 20 11 8 8 7 15 12 6 v4 20 ae 4 20 = me ie at 8 21 76 54 S 39 =* = = 10 21 ae 52 = a ss es a 16 22 a5 8 30 10 as =e == <= 23 49 oo = zs es =< == =. 24 41 = = a = at = =: 25 25 35 ae 30 =- es = 50 25 == 35 =a ae = = = 50 26 6 — =e = == — ie ae a we ; 1920] BIGELOW: REPORT ON CANNED VEGETABLES 465 TaBLe 1.—Concluded. SAMPLE AMPL: SAMPLE AMP ANALYST ae ree E 7 s a LE Ss ere SAMELE Shee 27 65 7 26 49 60 35 12 25 28 8 16 66 36 86 28 6 6 28 =e Je ~ se 23 ae 8 = 29 72 47 63 52 62 34 — 29 29 af 46° 37 = a = s 28 29A 29 38 52 50 42 34 8 28 29A =s 38 36 ee = == == 24 30 29 == = i i2 = 28 = 31 if 80 as 50 3 re as 46 31 ag 60 ey 22 aS =2 ee 20 32 80 48 = 46 aS as =f 36 32 =e 32 fe = =e =o =s 10 32A ts 44 as 49 as » is 20 32A ig 36 Ae =A =a eee 2 20 33 36 se a a3 Le = Se =? . 334 52 52 = =2 =: as ee =e i 34 70 64 a Be a Es be E: 35 35 38 as 44 ae = 2s 10 35 se 36 28 Ee ae — = 12 ’ 36 79 ai =e & oe = ee a 36A = 61 ad 58 = Ee 32 47 36A 2s 61 = a oe 3 ay 43 36B — 61 aie 55 = ae a2 51 36B 2s 48 we = 22 = = 42 36C 22 63 = 58 se Le ae 60 36C as 51 £ a oe ae < 48 37 9 60 60 70 50 68 32 40 38 23 100 se a2 33 ” = a2 Correct * By correct count is meant the count which it is believed will be obtained by the careful application of the Howard method. It is the average count obtained by six analysts who appeared to be most conversant with the method. 466 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol III, No. 4 TABLE 2. Bacterial count of tomato pulps. (Expressed as million bacteria per cc.) Se A ee ANALYST SAMPLE as: a a aaten Bae Sep aera a 1 32 17 28 31 45 30 ee 20 1A 33 20 22 32 57 409 | 12 30 2 40 40 19 46 85 500 3 39 3 34 15 72 22 62 240 3 17 3A 22 17 69 23 54 216 4 15 3B 26 15 65 26 52 176. |W 423 11 3B c 19 es pe a ALS gaat £6 4 26 24 24 40 86 144 6 12 4 re) 21 29 ee Pe: Bay 2 38 5 17 22 22 10 26 ba) | oa 19 5 es] 12 26 Us be Cte Oa 17 5A 15 31 17 14 31 58 | 12 19 +] 5A a 8 19 in ie Be: 32 6 17 26 23 27 23 37.| 22 42 q 7 48 20 40 40 100 | 1000 | 25 40 8 34 16 37 38 49 178 12 20 } 8 eA 29 38 a is ee 26 9 12 36 96 60 53 120 | 29 60 10 84 31 26 32 69 o71 | 49 40 10 aS 30 69 as Fe jal 40 1l 76 as Ri BY de Rags V3} Ne 12 86 28 50 59 43 160 7 27 12 x. 37 24 ee Bs RAG De: 30 13 27 12 60 20 60 125 | 12 20 13 E 20 20 ae i; qi 16 14 56 115 530 168 176 362 | 108 | 101 14A 29 23 91 14 19 35 7 13 15 31 21 24 29 50 216 7 17 16 10 97 194 132 77 86 | 75 93 17 24 26 54 34 so 180 8 43 18 96 mi a id i ee he ISA 22 es es iB ie ee oe. sé 19 ae 20 67 76 96 136 | 21 69 19A ee 25 52 61 57 181 | 108 42 19B ig 59 97 191 73 72 | 26 80 20 36 64 41 36 113 86 | 14 36 20 OE 38 37 ig “2 ar ee 31 21 168 a a 139 ie 52 ene 19 21 Bee 34 ae ae ae Gui Be 115 22 bi 8 72 19 id weg | a 23 43 =~ te se aia sates = =e 24 126 E be Be am Le 2 ae eae be 25 30 30 is 44 93 Bray || ie 38 25 aS 30 == = oe uae Se 36 26 24 - iy ae a ey.) Wie = 27 6 13 14 72 31 240 | 13 5 1920) BIGELOW: REPORT ON CANNED VEGETABLES 467 TaBLe 2.—Concluded. ANALYST are aba SAMPLE Si ee Sey Seyiies Ses era SA i 28 50 43 283 172 132 216 96 158 29 48 12 19 18 55 115 67 29 pe 13 48 —_ a= —_ = 96 294 40 Ss 24 16 34 96 14 57 294 = 12 49 es Ea zs a 96 30 40 = Ke = oe Be ss ie 31 Ly 15 = 62 = =a5 as 53 31 = 5 22 ome Re a pee ee 21 32 46 24 ae oo = aes >s. 44 32 a 35 = =. =e = a 44 324 a 23 ax 32 ze oe ae 30 32A 23 25 = == = eS a 43 33 if nas ay 2 a =a aa ee 338A 10 eS BS oe 7 cass a4 Z# 34 67 31 rx a == ES a =e 35 ile 36 = 36 = a! ae 104 35 == 48 = = 25 = = 38 36 61 = _ we as es nS ae 364A ae 41 oe 61 ue a. = 43 36A ee 29 = Ee 25 ee ue 50 36B =e 35 = 50 = — a 49 36B ae 30 Z =a 2 = = 34 36C 2 38 and 67 ae me - 41 36C es 44 a ae Ee ce Se =e 67 37 10 18 3 18 31 34 10 16 38 ue 34 me ual EY LES ae = Correct count® 32 17 50 30 af) 250 8 20 * By correct count is meant the count which it is believed will be obtained by the careful application of the Howard method. It is the average count obtained by six analysts who appeared to be most conversant with the method. 468 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE 3. Yeast and spore count of tomato pulps. (Expressed as number of yeasts and spores per ,\; cmm.) ANALYST SA Ae mar ws SA ks Sa a ae Ge veiire serge a 1 30 28 15 12 75 10 22 17 1A 22 26 12 12 71 13 13 14 2 38 54 20 20 87 20 7 46 3 28 30 18 15 72 20 6 18 3A 20 35 24 20 64 36 7 20 3B 26 30 28 22 70 10 =a 10 3B =s 34 ae =~ =e == = aes 4 34 21 21 30 148 20 8 20 4 a8 64 21 we == ae a 33 5 13 40 22 55 60 38 a 135 5 ae 121 12 ae me ee se 5A 12 40 29 50 50 41 16 133 5A =A 131 17 as =: == a2 70 6 16 40 16 48 18 30 18 28 7 110 43 29 3 83 =a if 8 8 32 35 10 10 46 18 4 18 8 = 26 12 = a == =s 24 9 12 44 28 38 36 30 24 32 10 18 20 10 14 96 28 6 32 10 = 10 29 as Be 2 = 12 11 32 oe = =e = =2 = =e 12 35 376 30 27 49 23 8 130 12 ut Bas 31 — =3 = we 53 13 43 12 9 12 54 6 6 24 13 ae 20 5 = ae = ae 16 14 62 166 142 27 65 88 21 40 144 142 200 147 83 100 45 30 43 15 36 20 12 11 56 14 8 21 16 2 26 5 5 34 5 {f 1 17 2s 62 25 32 130 22 9 33 18 20 =e me a a == =e ss ISA 106 = =e ze =e be = na 19 == 82 29 52 69 76 22 112 194 = 101 85 127 99 154 98 255 19B = 138 58 130 96 84 33 150 20 28 57 18 28 114 28 12 33 20 - 49 12 = oe aim a 22 21 34 70 == 30 ze 8 oe 10 21 + 20 = Be ae ee == 36 22 2: 22 119 23 = as a = 23 60 a oe os =e a5 Se = 24 6 ie ot 35 ae a= ae aS 25 18 20 == 20 = uy oe 24 25 as 18 = == Be = Le 18 26 5 se a Ss = 3s us == 1920) BIGELOW: REPORT ON CANNED VEGETABLES 469 Tasie 3.—Concluded. ANALYST a Serpe wo pert cred Sores Ge a ye Zl 25 175 2 16 110 20 20 120 28 4 0 0 0 0 0 29 25 39 8 14 61 9 4 12 29 = 13 8 ads fs an ae 12 294 30 34. 11 14 33 12 16 10 294 aye 16 27 = =5 = a= 14 30 30 = 23 as = mS = Ad 31 24 38 == 39 Ee =5 a 28 31 a 41 =o = = es “a 22 32 46 26 =a 13 —_ we ae 11 32 = 25 it == = Ee Ss 12 a SO PAN = 37 i& 17 = La oe 32 32A ea 33 =e = a 2 os 30 33 208 ae == = wes — “53 Ee 33A 30 =e ao = = » he ES Sy 34 84 112 = 23 = = = Be. 35 25 36 22 12 = ae _ 15 35 = 55 5 = é= oa ae 22 36 61 =a 25 ae ee =e a Le 36A Ee 31 as 19 f= a = 26 36A ica 38 a = = = mt 27 36B Se: 32 = =e == a 23 35 36B a 32 -s Bi == =? = 28 36C = 36 =; 30 = =e 2A 34 36C == 38 2 ae = = zs 32 37 1 850 500 200 48 16 10 14 38 aS 60 _— 25 aa =2 ae 3 Correct count® 30 30 18 15 72 20 10 il7/ » By correct count is meant the count which it is believed will be obtained by the careful Sypheston of the Howard method. It is the average count obtained by six analysts who appeared to be most _ conversant with the method. 470) ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE Microscopie count of samples taken at different stages of the SAMPLE ANALYST DETERMINATION NUM- SOLIDS | 1 j 7 BER 1 | TAS RI 2s a) Se Save 4 5 | | j SERIES 1. per cent 9 4.96 Molds} 2= 3Es. 2e= ae eee: 20 34 3 24 25 21 32 Bacteriat=4- = Ee ower 19 18 20 12 23 13 7 9a Yeasts and spores®_| _- sea if 9 5 2 12 9 7 10 5.88 Molds*232--s=25— ips Lie at Bs 14 26 30 18 19 28 Bacteniasa=s eae us Evie oe 4 oe 19 20 14 6 Yeasts and spores®_| ~~ eae ae 3 = 8 12 13 8 ll 7.34 Molds2)* 222225" oe ae ae 30 =f 34 22 24 a Bactemal= 22-322 ae nae 10 a 17 16 14 3 Yeasts and spores‘_| _- sete ae 4 ae 6 10 6 Le 12 | 10.58 Moldst2322 suns am SEL 30 40 40 36 31 26 42 IBactemals sa es ES oe 26 22 26 19 15 17 , 1 Yeasts and spores*_| __ aati 13 13 15 5 8 4 18 SERIES 2. 14 5.00 IM OlGgRA == oe Bere 8 12 10 20 16 9 17 102 Bactenats-e--- oe ae ae: 5 4 t 7 11 18 8 228 |) 256) abe Yeasts and spores®_| ~- = 5 2 3 6 8 a4 15 8.58 MoldS*222 == = ee os 12 = 24 32 Se _— Bacterial. sb a= =e Ges Be 5 ie 5 13 = we Yeasts and spores®_| _- ee ee 5 ie ll 24 =e as 16 15.22 MoldS* 23 Ae eee as ee 26 20 ae 26 24 a 33 Bacteria>...------ ae Ss 19 18 ~ 34 28 15 22 264 Yeasts and spores®_| -- Behe 14 8 oe 14 10 3 yy ® Expressed as per cent of microscopic fields showing molds. b Expressed as million bacteria per ce. 1920] BIGELOW: REPORT ON CANNED VEGETABLES A471 4, concentration of individual baiches of tomato pulp. ANALYST ] 7 AVERAGE 5A | 6 | 8 | 10 12 )\| “18 | 14 | 14A | 15 SERIES 1. | | 28 384 | 24 24 | 28 | 30 30 28 ee 27 3 ga 21 12 10 15 5 vi ya 3 34 8 224 5 5 pi 4 8 6 Ld 7 ae 244 22 =e = 26 40 26 am 25 Be. 274 10 Es se 12 3 5 om 10 = 2354 Zs = - 10 4 6 = 8 33 33 22 = oe 34 40 34 = 30 14 a4 iat a a 10 10 6 aa 12 16 2 1104 an 3s 8 8 7 ae 8 40 264 26 29 26 36 ae — % 34 ee 164 13. 10 14 19 a =4 es 17 7 22 BY 12 3 = 12 es Le _ 11 SERIES 2 13 30 12 13 6 6 10 32 12 13 124 124 5 152 8 9 5 14 5 4 10 8 174 244 12 164 1 1 10 2 6 12 3 5 5a 4d 19 362 12 ay 4 14 16 24 2 17 12 364 18 a 8 10 4 3 24 10 6 802 6 ae 4 5 8 3 4 26 284 10 * 10 24 22 34 20 22 224 19 464 22 ie 6 32 4 12 24 20 302 14 105? 6 ae 33 iM 10 ai 13 10 8 © Expressed as number of yeasts and spores per ;\; emm. ° Not included in average. 472 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE SAMPLE ANALYST DETERMINATION NUM- SOLIDS BER 1 Te ie eee sh | 4-2 ae SERIES 3 per cent | 21 5.03 Moldsss==sasnaces a Sper 49 60 51 56 46 pa 47 2 Bacteriate. =) re crane 7 5 13 14 14 12" 15 14 Yeasts and spores®_| __ ee 15 11 11 20 16 13 20 18 41 6.12 Molds*=222 82s 8 easel on 42 af 56 44 259 a Bacteria?! cen tes fe eee ees 10 = 14 7 7) = Yeasts and spores®_| __ ee Re 9 pe 20 20 17 oe 42 8.81 Molds es ae eae Les 54 ho 54 54 234 | 60 Bacteria? == === Pid Sekt Td 7 = 17 12 ile 7 Yeasts and spores*_| __ Emel = 10 2 18 20 23 21 25 13.17 IMGIO SS a ae = fate? i 42 us 55 46 184 | 59 Bacteria>*22- 22 =e Spend moi 12 =f 29 22 14 25 Yeasts and spores*_| __ ee =s 23 us 32 26 24 36 Series 4 | 47 | 5.47 IMolds*.2sce fase a ee 3 53 ed 58 57 52 ee 584 Bactertavss =. sea = o2e5 = 14 25 14 17 24 =z 7 Yeasts and spores*_| __ ee ty 7 ae 22 16 12 a 5d 22 6.35 Molds®- sees as ae 66 63 71 60 56 48 68 S44 434 Bacteriaesae 2-2 aes ae weet ih 11 12 10 8 13 19 7 od Yeasts and spores*_| __ Bett 10 3 7 9 12 18 23 gt lid 45 8.39 Molds? ew He ee Fe #2 63 =e 62 58 ze = 56" Bacteria>_...-___- A re ned 11 z= 14 14 = as 7 Yeasts and spores*’_| __ a=" a 13 a2 24 16 = a gd 46 9.70 Molds*-s=s<-2e =: = se ae 66 - 68 64 41 78 724 904 Bacteria®_...___.. os ah bs 13 = 24 29 14 18 Yeasts and spores®_| __ Sao — fi Bs 46 16 21 35 ® Expressed as per cent of microscopic fields showing molds. © Expressed as million bacteria per ce. 1920) BIGELOW: REPORT ON CANNED VEGETABLES 473 4.—Continued. ANALYST AVERAGE 5A | Guus 10 12 13 | 14 14A 15 SERIES 3. 52 424 36 66 64 52 66 60 46 53 584 72) 184 18 12 10 15 4 5 24 13 244 18 164 9 8 10 8 10 13 7 13 164 58 284 46 ae aS 64 50 52 ae 52 12 264 ae a a 12 4 5 as 10 1155 754 ine a = 10 12 16 23 15 = 514 36 =e 144 66 40 44 os 51 =e 454 29 Ei 13 8 3 4 aS 12 7 484 16 ee 10 14 9 12 ce 15 60 284 34 65 304 60 50 62 ae 54 264 17 654 27 21 24 24 7 6 Les 19 369 30 10002 28 8 6 13 49 50 a 27 184 SERIES 4 72 584 52 73 204 64 64 66 i 60 5 384 28 15 134 24 12 25 ae 18 11 554 9 10 Qa 7 20 31 aS 15 71 762 42 78 224 ee 76 42 68 64 60 Liz 602 24 28 64 a 8 4 17 14 5 16 744 12 8 104 10 24 10 12 12 gd 75 764 54 78 282 72 68 60 56 65 9 244 35 13 54 12 2 2 12 14 10 1284 14 10 ga 10 16 20 6 14 86 784 40 87 469 64 = = 48 66 884 624 19 324 10 16 14¢ 20 as oo 14 18 404 36 4004 12 12 102 15 ee = Uf 22 124 ° Expressed 4 Not included in average. as number of yeasts and spores per jy cmm. 474 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 4 TABLE SAMPLE, ANALYST DETERMINATION NUM- SOLIDS BER L(y .aA 2 3 | sa | < [is SERIES 5. | per cent 26 5.33 NMolds332 tee. ae —35hty a 6 ©: 20 22 18 20 Bacteria>_ #2. _ 2. 3 ae a 62 5 S4 89 | 2104 | 100 794 914 364 Yeasts and spores®_| __ ete ue 1 22 1 Y 5 20 29 12.98 Molds* 22. 272ee! aut skh Le 16 ae 24 14 8 16 Bacteriaree ce eke zs 3 See 2. 172 = 437 270 | 5044 | 226 1254 Yeasts and spores®_| __ Bee oe 6 = 14 6 12 6 SERIES 6. ' 31 17.20 | Moldssai 2) eee oe Boe oe 10 ee 16 12 ay 10 Bacteria> saunas see ee ppattle -. | 442 -. | 724 | 435 -- | 840 4008 Yeasts and spores®_| __ Bee bat 18 za 14 8 a 20 27 | 22.60 Molds. Seen. a eee a3 14 14 22 3 as 40 bacteriabeee see oe ones 22) 13662) 821 | 720) |U7508 -- | 880 6362 Yeasts and spores®_| __ —— _ 12 6 10 ga = 28 SERIES 7 28 5.00 Moldst 322228. 532 - a = 10 af 22 20 34 | 28 Bacteria>. _.- -- 2 - ae eet 2 357 .. | 451 | 415 | 9902 | 420 3384 Yeasts and spores®_| —- aoe ne 2 _- t 6 49) 15 32 8.42 Moluge se) ose re wae A 10 8 10 10 ze 18 Bacteria>_______-- ce ee a e) ee03 Ayes feyat.0) £S |Erou 6154 9608 Yeasts and spores®_| _- uses ae, 6 = 4 18 wt 24 33 | 14.07 Molds#2-).- -22 2 ae eae 4 8 4e 12 13 ee 12 Bacteria>___.____- an : ae 7208 -. |1656 {1620 -. }1000 765" Yeasts and spores®_| __ Jae é 1 aa 6 15 aa 24 34 | 17.83 Moldstu ee ee a OX? : 1S L 18 13 = 18 Bacteria®_____-_- aes sone Se LLS0) .- |1857 {1700 -- |1500 35204 Yeasts and spores*_| —- wee a} 9 =" 8 12 = 604 35 21.04 Moldse-=) Se ae oe = 8 E 6 12 ie 22 acteriayss 2s = eS 1155 _ |2381 |1768 -- {1500 15008 1900¢ |1486¢ Yeasts and spores®_| _- fs 18 32 6 8 a 674 ® Expressed as per cent of microscopic fields showing molds. » Expressed as million bacteria per cc. 1920) BIGELOW: REPORT ON CANNED VEGETABLES 475 4.—Continued. ANALYST 7 AVERAGE 5A 6 8 10 12 13 Uae 14A |e man SeRIEs 5. 164 244 12 oe 12 18 26 22 12 17 164 110: 784 92 =e 67 72 84 64 98 81 424 1084 1208 144 244 4 As 5 4 4 5 12 6 4d ae 184 18 we 4 18 6 8 6 13 me 4802 266 — 233 400 281 248 252 279 Es 6004 28 = 8 4 10 14 4 10 SERIES 6 = 102 r ae a 22 6 8 au 12 a 2408 += = =e 720 500 524 es 598 960¢ ee 1754 o- Be <2 2 8 12 a 12 30 124 te ce Bs 1 16 18 8 20 2884 6984 a2 aS iid 720 580 613 716 714 6008 444 2: Je a 4 8 13 @ ain SERIES 7. re 204 16 BS c= 14 10 10 84 16 Zs 2404 372 ze ze 480 3124 2954 3244 416 £2 144 2 mS Be: 1 17 16 5¢ 9 a 18 ze ae Se 16 8 8 ae 12 aa 1104 ae EA we 720 700 672 a 701 Le 1802 ae — Us 2 15 19 ES 13 == ae ee = £2 16 12 10 5 12 3 wa pies es se 20004 | 1125 1636 a 1407 AS x ee =e eS 4 19 16 =e 13 ‘ i 12 =e tn —— 10 12 14 a 14 &* ios 7504 ae =3 = 1800 2097 2092 me 1747 = 2004 =e —_ ae 2 8 720 xa 7 = 304 a a es 10 04 02 rs 12 a 7604 oe 22 2. 2400 2731 2477 a 2130 27712 =| 30334 == 4804 af oe Ae 4 8 8 = 9 © Expressed as number of yeasts and spores per ;/5 cmm. * Not included in average. ‘ Pi 476 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE ANALYST DETERMINATION SAMPLE! SOLIDS NUM- | BER 1 1A y 3 | 3A | 4 5 | SERIES 8. | per cent 37 4.88 Molds=23.- = tee at eek: AY 18 = 16 12 we = Bacteria>_________ 2 aes 2= 2025 ou 979 | 714 ous 8 6974 Yeasts and spores*_| __ sg ae 12 = 6 15 ae Sy 38 | 11.70 Molds*¢2 25 2 lea ce woes x 10 52 16 13 aks eS 1180 Bacteria>_________ EN te eS e0OE I! 2 748) 1485.1) ee : Yeasts and spores*_| __ = = 12 es 8 12 EL “ 39 | 20.10 Moldss=e2 222 ae 2A gee Ls 2s 14 ae 20 20 ai 23 Bacterial! 2222422 ae eae Spal -. |2477 |2160 a = 14504 Yeasts and spores®_| __ Sea us 12 =a 6 12 ae ae 40 | 28.80 IMolds*se eee se ee we 14 wa 15 a3 — Bacteria>_________ = ee =. 12095 -- |2592 |3000 = — 17504 Yeasts and spores®_| __ ee ae 15 os 12 18 ae = * Expressed as per cent of microscopic fields showing molds. ® Expressed as million bacteria per cc. i TABLE Bacterial count of pulp samples in Table 4 (Expressed as million ANALYST SAMPLE NUMBER 1 | 1A | 2 | 3 3A | 4 | 5 | 5A Series 1 9 32 30 34 20 39 22 12 5 10 Sa 6 ae 27 28 20 9 == 11 ae 11 ae 19 18 16 — 16 12 21 17 21 15 12 13 cf 9 SERIES 2. 14 8 7 7 12 18 30 13 8 15 a 5 ae 5 13 xe 12 oe 16 10 10 ss 19 15 8 12 10 1920) BIGELOW: REPORT ON CANNED VEGETABLES 477 4.—Concluded. ANALYST ] ] AVERAGE 5A | 6 | 8 | 10 | 12 13 | 14 | 144A | 15 SERIES 8. 6 20 boa =e ie 12 8 20 oe 14 1000 2304 a. =o as 875 880 1070 a= 935 40 4504 ao aa = 3 18 18 == 16 11 17 8 Ze st 36 8 6 =o 15 16 1100 4802 | 1600 =e we 960 1134 1478 a 1298 32 804 18 == £- 3 20 24 ae 16 6 15 ay a Es 14 10 16 ae 14 1000¢ 3504 ss ses ae 2400 2328 1931 ae 2170 44 2254 = ats == 4 16 12 = 15 = 24 10 = eel 20 10 6 = 13 ass 8004 | 4200 oe -- 3000 3888 2688 = 3066 e. 1604 12 — ae 6 3204 116 = 13 © Expressed as number of yeasts and spores per .4; cmm. 4 Not included in average. : 5. calculated to pulp of 8.37 per cent solids. bacteria per cc.) ANALYST AVERAGE 478 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE ANALYST SAMPLE NUMBER 1 1A 2 | 3 3A 4 | 5 | BA SERIES 3. 21 ipa 8 21 23 23 20 25 37 41 an 14 ss 19 10 23 Zs 16 42 es vy ae 16 11 16 7 ps 25 =85 8 ae 18 14 9 16 11 SERIES 4. f 47 his 21 els 21 26 37 ie 8 : 22 9 6 16 13 11 16 25 Ze 45 Ls 11 aes 14 14 = 9 = 46 = il BS! 21 25 12 16 16 SeRIEs 5 26 “2 97 a4 132 140 ee 157 = 29 As 110 ee 280 174 =f 145 ae SERIES 6. 31 at 215 ms | 352 210 =4 27 Ber ul aseds Cail 304 267 es SERIES 7. 28 ee 597 =8 755 694 a 32 se 699 = 710 696 =a 33 es arr = 985 964 ee 34 =e 554 mi 872 798 == 35 a 657 sf 946 702 = SERIES 8 37 ae 1758 me 1679 1223 = 38 a! 845 Ze 1252 1061 == 39 oe 716 #2 1030 899 as 40 a: 608 es 754 872 a * Not included in average. 1920 BIGELOW: REPORT ON CANNED VEGETABLES 479 5.—Concluded. ANALYST AVERAGE 8 | 10 12 | 13 | 14 | 14A | 15 SERIES 3. 32 20 17 25 7 8 a 21 a as a 16 5 ¢f = 14 28 = 12 8 3 4 = 12 17 13 15 15 4 4 aay 12 Series 4. 40 23 = 37 18 38 = 27 32 37 ss 16 11 5 as 17 30 15 = 12 2 2 Bes 13 9 14 ae 17 e = ef 16 Series 5. 144 ad 105 113 130 101 154 127 171 = 150 258 181 160 162 180 SERIES 6. —e = ss 350 244 255 se 292 ¥ ax Se 267 215 265 = 269 SERIES 7. se aa aS 802 5228 4948 == 712 = ake Se 716 697 668 ‘= 697 =e = a. 11908 669 974 22 837 = as = 844 984 982 = 819 =o ae a 954 1087 955 ak 846 SeRIEs 8. = 2s =f 1499 1504 1835 = 1601 ay ae SF 686 812 1040 af 954 = a Js 1000 969 804 a5 904 ae zs a5 872 1129 781 £3 890 ® Not included in average. 480 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE Microscopic count of trimming-stock pulp prepared from raw DETERMINATION Molds*22222==— Bacteriaes_--__ - 2 Moldg® 32. 222254 Bactena’y === Yeasts and spores® Molds®_____-__-- Bacteria>________ ® Expressed as per cent of microscopic fields showing mold. » Expressed as million bacteria per ce. ROT IN TOMATOES Less than 1.00 ¢ Expressed as number of yeasts and spores per « cmm. ANALYST 1A | 3 3A | 61 76 70 14 24 26 10 24 28 48 52 48 5 12 8 124 5 12 10 86 80 72 38 56 41 15 862 | 464 52 50 52 38 19 16 11 28 22 46 59 54 41 31 22 17 26 20 80 61 74 42 Bl 28 18 21 28 68 69 58 14 29 13 19 28 18 44 32 44 22 29 14 7 18 12 34 24 25 18 12 23 9 2 12 1920} BIGELOW: REPORT ON CANNED VEGETABLES 481 6. produci containing known amounts of rotting material. | ANALYST | | l AVERAGE 5A | 6 | 8 | 10 12 13 | 14 14A | 15 56 584 44 53 244 58 70 64 44 58 12 284 18 22 26 ae 4 11 26 18 5 17 6002 18 9 = 12 12 16 14 17 == 784 38 47 284 44 52 644 56 47 304 684 22 18¢ 6 5 4 5 5 | 3 10 8 es 494 9 5 22 13 8 10 8 ll A == = 78 92 ae ae == 2s 83 =e =< _ 29 36 =s se = ion 43 == ee us 13 20 22 = ae =e 19 52 = = 53 46 as a5 ax = 51 14 a5 = 24 20 = ae 23 ae 22 32 =e = 14 18 22 == sé ae 20 Ls = =e 65 50 = ss = ao 54 _ = as 31 30 = = eS = 29 =e we = 16 18 te we = = 21 82 984 464 84 56 86 74 70 64 73 804 742 944 28 324 19 16 13 28 63 54 40 28 24 544 14 18 13 30 354 424 11 20 69 854 42 79 284 56 324 oe 52 59 §4 354 14 21 10 408 22 =e 34 21 48 24 394 16 16 12 16 9 a 12 17 = 704 38 44 16¢ 44 36 =e 28 39 =e 444 26 18 84 1254 7 = 34 24 = 644 8 9 4 10 9 =e 14 ll 28 384 24 24 28 30 30 28 = 27 3 94 21 12 10 15 5 7 ES 13 8 224 5 5 oe 4 8 6 ae 7 4 Not included in average. 5 s © This figure is the average per cent of rot during a period when a number of batches of pulp were run. Therefore it does not accurately represent the amount of rot in the tomatoes for any individual batch. 482 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE Influence of delay in manufacture ANALYST DETERMINATION pees STORAGE 1A | 3 | 3A | 4 i hours 26 2 Molds*22 22) sa 2 eee ae aa 6 20 22 18 Bacterial: =e acres es nfs 62 84 89 2104 794 gid Yeasts and spores®__--_- fe dis 1 1 4 5 28 8 Molds8))~ %22 ay aes ee ili 10 22 20 38 Bacterial 22 8a oe bee Ba as 357 451 415 9904 3384 Yeasts and spores°_-____- a Bi 2 4 6 4a 36 12: Molds8 2) Fao- 22 es ae ae 10 22 16 2 IBacteriabaee =e eee ue ee 396 874 520 990 4124 8424 Yeasts and spores®_-____- a ES 2 10 15 4 ® Expressed as number of microscopic fields showing mold. b Expressed as million bacteria per cc. 1920 BIGELOW: REPORT ON CANNED VEGETABLES 483 te on microscopic count. ANALYST rl AVERAGE 20 164 244 12 12 18 26 22 12 17 100 1102 784 92 67 72 84 64 98 $1 364 424 1084 20 143 244 4. 5 4 4 5 12 6 28 me 204 16 = 14 10 10 84 16 420 at 2409 | 372 =2 480 3124 2954 3244 416 5364 | 3204 165 ae 143 2 at 1 HG, 16 ite 9 6 == 204 6 = 14 18 24 8 13 420 =A 2404 509 =< 480 310 487 324 531 15 — 144 14 = 7 14 10 5 10 e Expressed as number of yeasts and spores per \; cmm. 4 Not included in average. i 484 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 4 TABLE Influence of finishing on the ANALYST DETERMINATION Reston DESCRIPTION 1 | 1A 2 3 | 3A Series 1. 23 Unfinished IMolds*=s-s ee 2 = Bret D 60 40 42 49 46 584 Bactenaleses ier bos hs i 19 20 20 19 22 174 Yeasts and spores®__________ Ee 2 ee 12 7 8 20 26 48 Finished Molds8 =o ind on ae ae Lay 38 48 ue 46 53 Bacteriat’:- 220 ee a Lee 22 28 AS 29 16 Yeasts and spores*__________ os 2 13 8 a3 20 18 Series 2. 17 Unfinished Molds®= 332 ee eee ae aoe 28 20 =e 22 32 Bacterial i ares ues eae = AS 10 14 as 29 26 Yeasts and spores®__________ a RS 9 15 ~~ 34 28 18 Finished Moldstioe2 0) ee eee 2 seus 28 36 i= 26 30 iBacteriabh==<2 2" aw eaeae S.. zee 13 18 = 24 29 Yeasts and spores®__________ ee prt 18 ll 22, 14 24 SERIES 3. 43 Unfinished Molds: 2.2.22 ee. Ae Se ee 28 28 aH 27 31 414 Bactenabe= si 2 23 oes. Seen a eo 10 16 a 34 23 Yeasts and spores*__________ as sone 20 5 ss 22 18 44 Finished IMoldbs!=2)32 ro eee Ee es 28 28 & 29 28 pacteriaks 522 eee sees oe = 15 15 = 38 23 Yeasts and spores*_____--__- Ze Bote 11 9 ioe 21 28 * Expressed as number of microscopic fields showing mold. » Expressed as million bacteria per ce. +. 1920) BIGELOW: REPORT ON CANNED VEGETABLES 485 8. microscopic count of tomato pulp. ANALYST AVERAGE 4 | 5 | 5A | 6 | 8 | 10 | 12 | 13 | 14 | 14A | 15 Series 1. 16 42 39 864 a | 50 44 ae a = 54 43 4 26 14 15 ASI |= = 13 14 18 = =e 36 20 zs 594 | 634 602 ‘ 174 13 32 32 662 54 6 10 20 5s ee 5 18 144 31 54 48 802 | 20 = 20 40 =e = Bor 43 66 24 23 36 33% | 21 = 24 28 =e at 36. 27 39 Parse 20),| see 14) Lapdog Tl a 9 14 SERIES 2. 8 tea Be 404 8 34 12 24 12 24 28 22 2 17 17 ae 484 14 15 10 28 4 11 26 17 LAS 604 9 22 ee Aca to 12 10 16 24 18 8 17 254 11 26 ae 352 | 10 37 14 30 10 24 44 25 16 16 ae 294 | 22 14 38 on 12 16 38 22 5 Poet ey ao ee) Rl ore. Mh sa. oa | 10 19 SERIES 3. 24 =f 20 264) 14 32 4 22 16 zt 34 23 25 ss 6 489 | 35 ll me 22 8 es 22 19 324 28 be 23 464) 13 5 10 10 15 ae 18 16 19 as 20 Eh | 20) 32 8 ee 28 a 28 24 2 19 7 20 344 || 12 4 13 30 4 Bt 24 23 42 16 2 16 402 10 11 6 8 16 ps 18 14 124 c Ex cr C Pamcmamacs. ee 486 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 4 REPORT ON COCOA AND COCOA PRODUCTS. By EuGcEeNE BioompBere! (Bureau of Chemistry, Food and Drug Inspec- tion Station, Buffalo, N. Y.), Associate Referee. The name “Cocoa and Cocoa Products” is rather inaccurate. Cocoa is merely one of the products made from the bean of the Theobroma cacao. Neither chocolate nor cacao butter is a cocoa product. All of these can properly be classified as “Cacao Products’. This nomenclature has been adopted by the Bureau of Chemistry in the information cards and by the Committee on Editing Methods of Analysis and should be adopted by the Association of Official Agricultural Chemists. MILK CHOCOLATE. At the 1914 meeting it was recommended that the associate referee on cocoa products for the year 1915 make a study of the manufacture of milk chocolate with a view to finding out whether different methods of manufacture might render the casein insoluble in the reagents used for its extraction in the examination of milk chocolate. No work was done in 1915 on this recommendation. At about that time the present asso- ciate referee had occasion to do some work on this subject. It was felt in some quarters that continued heating might have the effect of render- ing the casein in milk chocolate insoluble in the 1 per cent sodium oxalate solution which was used for its extraction. To ascertain whether or not this was true, milk chocolate of known milk content, milk powder and commercial casein were heated at a temperature of 60°C. (which is the highest temperature to which milk chocolate is heated in the process of manufacture), for a period of 192 consecutive hours, samples being taken from each every 24 hours. The samples taken from each product at each period were analyzed and the amount of casein found in each product did not vary over the 192-hour period. Evidently, heating these products did not render the casein contained therein insoluble in 1 per cent sodium oxalate solution. Samples were then collected by the associate referee in person from factories to ascertain whether casein was rendered insoluble by the various methods of manufacture. There are three principal methods in the manufacture of milk chocolate, the differences being in the way the milk is added: First, using fresh milk; second, using condensed milk; third, using milk powder. Samples were taken at various stages of manufacture from each concern and analyzed to ascertain whether by any steps in the manufacture casein might be rendered insoluble in the sodium oxalate solution. 1 Present address, P. O. Box 54, Station E, Cleveland, Ohio. 1920] BLOOMBERG: COCOA AND COCOA PRODUCTS 487 TABLE 1. Determinations of casein in milk chocolate al various stages of manufacture. | STAGE OF CASEIN | CASEIN METHOD OF MANUFACTURE | MANUFACTURE FOUND THEORETICAL Sample 1: per cent per cent Reon treshymn kee ee oe KS |e Mixed 4,24 4.14 Birompimeshy milks este ee tS ee Rolled 4.14 4.14 OEP BITE BRET TEIN UE soe a el an OP Conched 4.29 4.14 Sample 2: Rieommtreshim lies oe ee tS Mixed 4.24 4.32 OSTA aye NT ee ee er eee Rolled 4.28 4.32 eombiresh mikes ae 2 Che te ere eee Conched 4.24 4.32 Brom condenkeavmilk = =— =~ = === Mixed 1-22 1.12 Prom condensed milk = _ _ Vas se ee Finished 1.18 1.12 Bromennlkepowd ers= sa or ee Sa | Mixed 3.13 3.26 Rromimilks powderse%. 2 ao 8 = 8 2s bd Finished 3.10 3.26 The results obtained in this investigation would appear to show con- clusively that the process of manufacture does not render casein insolu- ble in the reagent used. During the course of this investigation some trouble was experienced in obtaining theoretical results for casein when analyzed by the Baier and Neumann method’, which was provisionally adopted by the Asso- ciation of Official Agricultural Chemists. It seemed that there were two operations in this method whereby errors might enter. In the first place, although the method calls for a thorough rubbing up of the defatted chocolate with the sodium oxalate, this material has a tendency to settle out, and it is possible that in some cases not all of the material present is acted upon by the solvent. This would have the effect of giving results which would be too low. On the other hand, there is a possibility that on the precipitation of the casein other bodies are carried down which might with difficulty be removed from the precipi- tate by washing, especially as this precipitate is rather sticky. Minor objections are that the heating in a 250 cc. flask had to be cautiously done to avoid foaming over and that the filtration of the sodium oxalate solution was very slow. To ascertain whether the defatting operation was necessary, casein was determined in samples both with and without previous defatting, and it was found that no necessity existed for defatting the sample; in fact, if anything, the heat dissolved the fat from the milk chocolate and left the chocolate material in a very finely divided condition, so that the action of the sodium oxalate was accelerated. In order to obviate the necessity of washing the acetic acid precipitate, 1Z. Nahr. Genussm., 1909, 18: 17. 488 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 the following modification of the Baier and Neumann method was worked out by the associate referee: Transfer 10 grams of grated milk chocolate to a 500 cc. Erlenmeyer flask. Add 250 cc. of 1% sodium oxalate solution, heat to boiling and boil for about 2 minutes. Allow to cool, add 5 grams of magnesium carbonate and filter. Determine nitrogen in 50 cc. of the filtrate, corresponding to 2 grams of the original chocolate. Pipette 100 ce. of the filtrate into a 200 cc. flask, make nearly to volume, then add 2 ce. of glacial acetic acid. Make to volume, shake, allow to stand for some time, so that the precipitate may settle somewhat, then filter. Determine nitrogen on 100 cc. of the filtrate, which also corresponds to 2 grams of chocolate. The difference in the amounts of nitrogen is the nitrogen which is present as casein. This, multiplied by 6.38, gives casein present in 2 grams of milk chocolate. TABLE 2. Casein determinations. (Analyst, Eugene Bloomberg.) BAIER AND epee Bue METHOD OF MANUFACTURE OnE plead een see DEFATTED beatae MODIFICATION per cent per cenl per cent per cent Made from condensed milk _____ 1.12 eh 0.80 1.16 Made from condensed milk _____ 2.31 1.76 1.52 a ies bet en 2.32 very site ae 2.36 Made from condensed milk - -___ 1.09 wake 0.57 a Made from whole milk_________ 4.14 3.80 3.92 4.14 fee 3.94 4.12 4.24 Beit: 4.18 4.24 4.29 Sanh 4.31 4.42 sous Made from whole milk_________ 4.32 4.58 4.12 4.20 aee2 4.90 4.28 4.24 3%) oye 4.38 4.24 Made from milk powder________ 3.50 aLue aie se 3.17 wee aod ms 3.26 er Be. be 3.34 asco Sead ee 3.35 Made from milk powder_______- 3.26 = es oho 3.13 edz es sake 3.13 28 Bi ta 3.17 Made from milk powder________ 3.58 oes par 3.53 ps. met Pusat 3.57 Be ol. nee Pet 3.70 Made from milk powder________ 4.48 aero Sout 4.59 ere aoe! eel 4.59 Made from milk powder_______- 3.50 wat eas ae Je mae vee 6 1920] BLOOMBERG: COCOA AND COCOA PRODUCTS 489 A large number of samples of known milk content were examined by this method and also by the Baier and Neumann method, with results as given in Table 2. These samples were all made under the direction of the associate referee, the milk used being analyzed and the casein calculated as 80 per cent of the milk protein. It will be noted that, where there is a considerable amount of casein present, the Baier and Neumann method, either with or without defat- ting, gives results very close to the theoretical. However, where there is but a small amount of casein present, this method tends to give results which are considerably low. The proposed method, however, gives results which are very close to the theoretical, irrespective of the amount of casein present. COOPERATIVE WORK. One sample of milk chocolate was sent to eleven collaborators and reports were received from nine. The following directions were sent out: Fat.—Determine fat on a 2-gram sample, using anhydrous ethyl] ether, extracting in a Knorr or similar condenser for 4 hours. Evaporate off the ether, weigh the ether extract and repeat the extraction for another 4 hours, reporting results obtained at the end of 4 hours; also results obtained at the end of 8 hours. Also report if there seems to be any theobromin in the residue of either extract. Casein.—Determine casein by the method given in the 1912 Proceedings'. It is unnecessary to add uranium acetate in the precipitation and in washing of the pre- cipitate. Also determine casein by the following method: (It is unnecessary to defat the chocolate.) Weigh 10 grams of chocolate into a 500 cc. Erlenmeyer flask and add 250 cc. of 1% sodium oxalate solution. Heat to boiling and boil gently for a few minutes; then cool, add 5 grams of magnesium carbonate and filter. Determine nitrogen on 50 ce. of this filtrate, corresponding to 2 grams of the original sample. Pipette 100 cc. of the filtrate into a 200 cc. volumetric flask and make almost to the mark with water. Then precipitate the casein by the addition of 2 cc. of glacial acetic acid. Make to volume, shake, filter, and determine nitrogen on 100 ce. of the filtrate. The difference in the two nitrogen determinations gives nitrogen derived from the casein, which, multiplied by 6.38, will give the amount of casein present in 2 grams of the sample. Report which method you consider the better from the standpoint of time and convenience. Sucrose and lactose-—Determine sucrose and lactose by the polariscopic method* except that the present formule? are used in place of those there given. Determine lactose by copper by heating on a water bath 10 grams of the chocolate in a 250 ce. volumetric flask with approximately 200 cc. of water, shake occasionally so that all sugar is dissolved. Clarify with lead subacetate, make to mark, filter, remove the lead and determine lactose on 50 cc. aliquot. Report lactose as anhydrous lactose. 1U.S. Bur. Chem. Bull. 162: 130. 2 Thid., 137: 98. ® Assoc. Official Agr. Chemists, Methods, 1916, 32y-. 490 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE 3. Results reported by collaborators on milk chocolate. FAT FAT SUCROSE LACTOSE CASEIN Polari- Polari- Bai SNSE SESE eno bi a deonie Bonk Copper fui a Proposed tion tice. deter- deter- | method |Neumann| method mination | mination meth per cent | percent | per cent | per cent | percent | per cent | per cent J. Callaway, jr., U. S. | 33.06 | 33.06 | 39.33 | 8.19 7.68 3.52 3.30 Food and Drug Inspec- tion Station, U.S. Cus- tom House, Savannah Ga. W. C. Taber, Bureau of | 32.64 | 32.80 | 40.56 6.89 7.40 3.74 3.85 Chemistry, Washing- ton, D. C. W. L. Dubois, Hershey, | 32.82 | 33.12 | 40.00 | 6.56 ee 3.50 2.088 Pa. C. L. Black, U. S. Food | 32.84 | 32.84 | 39.92 | 6.90 7.14 4.15 4.03 and Drug Inspection | 32.80 | 32.84 | ___- mene eee 4.02 Seen Stations Us SAwpprais=i|p eee ellen ee sue sume 3:96P4)) eee er’s Stores, Philadel- phia, Pa. Leonard Feldstein, U. S 32.80 | 32.92 | 38.00] 5.902 7.00 ae 3.54 Food and Drug Inspec- | 32.70 | 32.81 | _-_- peda See pal 3.34 tion Station, Tabor Opera House Building, Denver, Colo. A. G. Woodman, Massa- | 32.93 | 33.08 | 40.05 | 6.41 a 3.57 Lie chusetts Institute of Technology, Boston, Mass. GiBayWarriner, Ue) S2 lipoo0re lo.2le) eee ans = .. 3.50 3.04% Food and Drug Inspec- tion Station, Territorial Board of Health, Hon- olulu, Hawaii. A. S. Wells, Dairy and | 33.03 | 33.56 | 43.81 8.18 7.81 3.565 | 3.74 Food Commission, | 33.01 | 33.52 | __-- pie 7.91 —— oo Portland, Ore. F. T. Anderson, U. S. | 31.56 | 31.56 | 40.33 | 6.60 7.39 3.80 3.76 Food and Drug Inspec- tion Station, U.S. Ap- praiser’s Stores, New York, N. Y. Eugene Bloomberg, U.S. | 32.75 | 32.85 | 40.01 6.75 7.59 AeA 4.16 Food and Drug Inspec- | 32.78 | 32.88 | ---- moh face aon shes tion Station, Federal Building, Buffalo, N.Y. Manimuni= 2222s aes 33.07 | 33.56 | 43.81 8.19 7.91 4.15 4.16 Minimise = So 31.56 | 31.56 | 39.33 6.41 7.00 3.50 3.30 Averageuno- 40 cae oe one 32.77 32.93 | 40.50 7.06 7.49 3.73 3.72 heoreticale. =.) Sa Bia ha 40.00 7.00 7.00 4.14 4.14 * Omitted from averages. mie » Not defatted for casein determination. 1920) BLOOMBERG: COCOA AND COCOA PRODUCTS 491 Messrs. Callaway, Feldstein, Warriner and Anderson report theo- bromin present in both the 4- and 8-hour fat extractions; Messrs. Taber, Dubois, Woodman and Bloomberg report theobromin present in the second extraction only; Mr. Black reports no theobromin found. Messrs. Callaway, Warriner and Wells prefer the proposed method for casein determination; Messrs. Dubois and Black prefer the first. The others indicated no preference. DISCUSSION. Several of the collaborators called attention to the fact that while the instructions called for lactose by copper to be reported as anhydrous, the normal way in which lactose occurs, and the way it is calculated by the formula from the polariscopic readings, is with 1 molecule of water. This is correct, and lactose by copper should be reported as mono- hydrated. Calculating by this table, Mr. Taber found 6.89 per cent of lactose (monohydrated) present. Mr. Taber pointed out, however, that the presence of sucrose gives results for lactose by copper which are high, and a table should be used for this determination which takes into account the sucrose. The formula sent out for the polariscopic determination of sucrose and lactose was worked out by Messrs. Seeker, Shanley and Lourie of the U. S. Food and Drug Inspection Station of the Bureau of Chemistry, New York, N. Y. The formula as adopted by the association in 1910 calculates the amounts of sugars present and then calculates the increase of volume due to these sugars, the percentage of sugars then being corrected to account for this increased volume. This, while near enough for ordinary purposes, is not exactly correct. The formula sent out this year calculates the increase of volume directly from the polarization and is theoretically correct. It seems to be unnecessary to dry over sulphuric acid before extracting fat with ether. Mr. Callaway obtained 33.06 per cent of fat without drying, and 32.96 per cent after drying. Your referee obtained 32.78 and 32.75 per cent without drying, and 32.77 and 32.78 per cent after drying. The results obtained on the fat extraction seem to show con- clusively that a 4-hour extraction, using anhydrous ethyl ether, removes all the fat present in a product of this nature. Long extractions are not necessary on cacao products for the reason that in their manufacture they are ground very finely and for this reason the fat is easily extracted. Moreover, it appears that a longer continued extraction will extract a portion of the theobromin. There is a difference of opinion as to which casein method is the shorter and more easily carried out, although the results obtained are about the same by either method. The associate referee has made 492 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 several hundred determinations and prefers the new method, but there is no doubt that Mr. Black’s objection is well taken, that there is con- siderable frothing of the sodium oxalate solution when treated with sulphuric acid, due to the evolution of carbon monoxid. Several other solvents were tried, but in no case did they give as good results as the sodium oxalate. CACAO BUTTER SUBSTITUTES. The detection of cacao butter substitutes in cacao products offers some difficulty, especially if these substitutes are present in small amounts. The general method of procedure is to determine the con- stants on the extracted fat and to depend upon these determinations for the detection and identification of the cacao butter substitute used. If a substitute is present in amounts of from 5 to 10 per cent, it is with difficulty detected; in fact, some of the substitutes, especially tallow and hydrogenated cottonseed oil, have constants so near those of cacao butter that it is practically impossible to detect their presence by this procedure. It was desired to find a short and more exact method for the detec- tion of the presence of cacao butter substitutes. Such a method has been found in the critical temperature of dissolution of the fat in glacial acetic acid. For a given strength of glacial acetic acid the critical temperature of dissolution of a cacao butter is constant within 1°. This test is practically the Valenta test!. It is made by adding 5 ce. of the melted filtered fat to an equal volume of glacial acetic acid in a test tube, the whole being heated, with constant stirring, until the fat goes into solution. The solution is then allowed to cool (stirring with a thermometer) and the temperature at which turbidity appears is the critical temperature of dissolution. The critical temperature of disso- lution of any fat varies with the strength of acetic acid used. Inasmuch as slight differences in the strength of the acetic acid make considerable difference in the critical temperature of dissolution, it was thought advisable in this determination to standardize the acetic acid each time it was used against an authentic sample of cacao butter, rather than attempt to use an acetic acid of definite strength. Since this is the case, it is not the critical temperature of dissolution found which is indicative of the purity or adulteration of the sample, but rather the variation of the sample from an authentic sample of cacao butter. Practically all substitutes give a critical temperature of dissolution considerably below that of true cacao butter. Using an acetic acid with which pure cacao butter gave a critical temperature of dissolution of 96°C., shell butter gave 71°C., coconut and palm kernel products iJ. Soc. Chem. Ind., 1884, 3: 643. 1920 BLOOMBERG: COCOA AND COCOA PRODUCTS 493 16°C., hydrogenated cottonseed oil 104°C., tallow 94°C., corn oil 20°C., cottonseed oil 46°C., olive oil 63°C., peanut oil 52°C., sesame oil 40°C. Mixtures of any of these products with pure cacao butter changed the critical temperature of dissolution by an amount approximately propor- tional to the amount of substitute used. The only substitutes in use which have a critical temperature of dissolution equal to or greater than cacao butter are hydrogenated oils and tallow. These adulterants may be detected by adding 5 cc. of the melted fat to 5 cc. of a mixture of equal parts of acetone and carbon tetrachlorid and allowing the mixture to stand in ice-water. In the presence of either hydrogenated oil or tallow a flocculent precipitate will be formed in from 5 to 30 minutes, depending upon the amount of adulterant present. This test is simpler, much more exact, and will detect smaller additions of tallow than will Bjorklund’s ether test?. These two tests may be easily and quickly made and give an almost certain indication as to the presence or absence of cacao butter substi- tutes as well as some indication of the kind of substitute used. To ascertain the exact identity of the substitute, it is necessary to deter- mine some of the ordinary constants. If, however, the critical tempera- ture of dissolution is approximately that of cacao butter, and neither hydrogenated oil nor tallow is found, it is almost certain that the product under examination is a pure cacao butter. COLLABORATIVE WORK. The following samples were sent to collaborators: Sample 1.—A mixture of 80 per cent cacao butter and 20 per cent coconut oil stearin. Sample 2.—A mixture of 80 per cent cacao butter and 20 per cent cottonseed oil. Sample 3—A mixture of 80 per cent cacao butter and 20 per cent coconut oil stearin. (This sample is practically the same as No. 1, except that the coconut oil stearin is put out by a different manufacturer under a different name.) Sample 4.—A mixture of 80 per cent cacao butter, 10 per cent coconut oil and 10 per cent corn oil. Sample 5.—A mixture of 80 per cent cacao butter, 15 per cent coconut oil and 5 per cent tallow. Sample 6.—A mixture of 95 per cent cacao butter and 5 per cent coconut oil. Sample 7.—Pure cacao butter. There was also forwarded a sample of cacao butter which was to be used as a standard. INSTRUCTIONS TO COLLABORATORS. Make the critical temperature of dissolution determination and report results obtained on each sample, also those obtained on the authentic sample of cacao butter, using the same acetic acid on all tests. Make a test for tallow and hydrogenated oil, using 5 cc. of the melted fat and 5 cc. of acetone. Heat if necessary to dissolve, and 'Z. anal. Chem., 1804, 3: 233. 494 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 4 allow the mixture to stand in cold water overnight, running a blank with pure cacao butter at the same time. A white flocculent precipitate indicates the presence of either tallow or hydrogenated oil. Should either of these tests give results which would indicate the presence of a foreign fat, it is requested that other determinations, such as the saponification number and iodin number, be made in order to determine, if possible, the kind and percentage of adulterant in each sample. TABLE 4. Collaborative results on cacao butter substitutes. CRITICAL DIFFER- ANALYST non Rod ACETONE TEST BON {tio CONCLUSIONS ‘ j DISSOLU- CACAO ; t NUMBER ert! TION BUTTER SAMPLE 1. | 2G, °C. 1 71.0 —17.0 negative 203.4 28.4 | 20 percent coconut or palm kernel stearin present. 2 83.5 —13.0 slight ppt. 206.3 29.6 | 20 percent coconut or BES See as) a re ae eps 29.4 palm kernel stearin present. 3 81.5 —16.5 negative 204.2 30.5 | 20 per cent coconut or palm kernel stearin present. 4 95.5 —11.5 negative 205.3 29.2 | Adulterant present. pee | ame oh ies eda 206.3 29.3 5 75.5 —17.0 negative 205.5 29.2 SAMPLE 2. 1 75.0 —13.0 negative 194.2 47.3 | 20 per cent corn or palm oil. 2 85.0 —11.5 slight ppt. 195.3 50.2 | Largely shell butter. 3 87.0 —11.0 negative 194.8 51.8 | 20 percent cottonseed oil present. (Hal-— phen test positive.) 4 100.25 | — 6.75} negative 198.2 49.5 | Cottonseed oil pres- SEERA Sy RRR)! 7 Sed 199.0 49.4 ent. (Halphen test positive.) 5 81.5 —11.0 negative 198.2 40.9 SAMPLE 3. 1 70.5 —-17.5 negative 204.4 30.0 | 20 per cent palm or coconut stearin. 2 79.0 —17.5 negative 205.3 29.1 | 20-25 percent coconut Sats Besser all eeprers ses 207.0 29.4 or palm kernel oil. 3 80.0 —18.0 negative 205.5 30.3 | 20 bea cent coconut olein. 4 94.0 —13.0 negative 209.0 29.4 | Adulterant present. Sy Bote a ee SOE ae 208.4 29.4 5 82.0 | —10.5 | slight ppt. 203.9 | 29.5 pe See eee eee a ee a i " 1920) BLOOMBERG: COCOA AND COCOA PRODUCTS Tasie 4.—Continued. 495 CRITICAL DIFFER- | TEMPERA- ENCE | SAPONI- 5 ANALYST TURE OF FROM ACETONE TEST FICATION Wen CONCLUSIONS DISSOLU- CACAO NUMBER | NUMBER TION BUTTER SAMPLE 4. eco 1 71.5 —16.5 negative 198.8 41.5 2 81.0 —15.5 slight ppt. 201.9 40.8 | Largely beef tallow. ace eal) eee SSS 202.2 41.0 3 81.0 —17.0 negative 200.7 41.0 | Adulterant present. 4 96.25 | —10.75| negative 201.1 40.5 | Adulterant present. Soe) ee 6 } 200.1 40.3 5 74.5 —18.0 negative 199.3 40.5 SAMPLE 5 1 71.5 —16.5 positive 203.6 29.8 | 20 per cent hydrogen- | _ ated cottonseed oil. 2 81.5 —15.0 heavy floccu- | 205.9 29.9 | Tallow present. ee eee lent ppt. 206.3 29.9 3 82.5 —15.5 positive 204.4 30.0 | Tallow present. 4 95.0 —12.0 heavy ppt 203.0 29.5 ease!) || eee eee 202.3 29.7 5 74.0 —18.5 white floceu- | 206.2 29.8 lent ppt. | | SAMPLE 6. | 1 79.0 —9.0 negative 195.9 34.8 | Less than 10 per cent foreign fat. 2 91.0 —5.5 slight ppt. 197.9 34.0 | 5 per cent oleo stearin or palm or coconut product. 3 93.0 —5.0 negative 197.3 34.5 | Adulterant present. 4 105.0 —2.0 negative 198.6 33.6 | Adulterant present. wee eee ee 197.8 33.9 5 86.5 —6.0 negative 197.3 33.8 SAMPLE 7. 1 88.0 0.0 negative 192.3 36.0 | Pure cacao butter. 2 97.0 +0.5 negative 194.2 36.0 | Pure cacao butter. 3 101.0 +3.0 positive 193.7 36.5 | Hydrogenated cot- tonseed oil present. 4 108.0 +1.0 negative 194.5 35.8 | Pure cacao butter. 5 92.5 0.0 cloudy 195.2 35.8 496 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS |Vol. III, No. 4 Tasie 4.—Concluded. l CRITICAL DIFFER- TEMPERA- ENCE | SAPONI- IODIN ANALYST | TURE OF | FROM ACETONE TEST FICATION NUMBER CONCLUSIONS DISSOLU- | CACAO NUMBER TION | BUTTER | i | SAMPLE 8, STANDARD Cacao Butter. °G: 20; 1 88.0 Jute negative wae bie | | Da E96:5 LBS Ee| Pe ae 192.8 35.7 3 98.0 = est negative Lye tee = be 4 107.0 % 8 negative 192.7 35.8 a's SE Ma) COPS oS 192.1 Dat 5 92.5 a) eS Se 195.4 =e Analysts referred to in above table are: 1. W. C. Taber, Bureau of Chemistry, Washington, D. C. 2. W. W. Karnan, U. S. Food and Drug Inspection Station, U. S. Appraiser’s Stores, Boston, Mass. 3. J. Callaway, jr., U. S. Food and Drug Inspection Station, U. S. Custom House, Savannah, Ga. 4. C. L. Black, U. S. Food and Drug Inspection Station, U. S. Appraiser’s Stores, Philadelphia, Pa. 5. L. D. Elliott, U. S. Food and Drug Inspection Station, U. S. Appraiser’s Stores, New York, N. Y. DISCUSSION. On Samples 1 and 3 the low critical temperature of dissolution, the high saponification number and low iodin number are all indicative of the presence of a coconut oil product. On Sample 2 the low critical temperature of dissolution would indicate the presence of an adulterant and the high iodin number would show that this adulterant was one of the fixed oils. The only way in which it could be definitely recognized would be by making the specific tests, as was done by two of the analysts. On Sample 4 the low critical temperature of dissolution indicates the pres- ence of an adulterant. The high saponification number would indicate the presence of some coconut oil product, and the high iodin number would indicate that one of the fixed oils was also present. Inasmuch as there is no specific test for corn oil, it would be practically impossible to tell exactly what oil was there. Analyst No. 2 reported the presence of beef tallow on this sample, basing his conclusions on the saponification number and the iodin number. However, the critical temperature of dissolution would show conclusively that it was not tallow, as would the acetone test. In Sample 5 the acetone test would show conclusively 1920) BLOOMBERG: COCOA AND COCOA PRODUCTS 497 the presence of either tallow or hydrogenated cottonseed oil. At the same time, the low critical temperature of dissolution would show the presence of some other adulterant, and this, together with the high saponification number, is practically proof positive that this other adul- terant is coconut oil. In Sample 6 the low critical temperature of disso- lution and the slightly increased saponification number show the presence of a coconut oil. As several of the analysts suggested, it is essential that in the critical temperature of dissolution exactly 5 cc. of the melted fat and 5 cc. of acetic acid be used. It may be noted that although Samples 1 and 3 are practically identical, still some of the analysts obtained differences in the critical temperature of dissolution. This may be partly explained by the supposition that they were not careful to use exact amounts of fat and acid. The critical temperature of dissolution obtained by the different analysts on the same sample varied considerably. This is due to the fact that the acetic acid used varied in strength, and, as has been pointed out, a slight variation in the strength of the acetic acid makes a considerable variation in the critical temperature of dissolution. This also shows why it is important to standardize the acetic acid used against an authentic sample of cacao butter each time. In every case where an adulterant was present the critical temperature of dissolution was lowered sufficiently to indicate this fact. Several of the analysts reported that in the determination of the ace- tone test a part of the cacao butter solidified on standing, thus somewhat obscuring the flocculent precipitate. However, it will be noted that the flocculent precipitate thrown down when tallow is present is so distinctive that no one failed to note its presence in Sample 5. Moreover, in only one case was a positive test found when no adulterant was present. It was desired if possible to obviate this difficulty and to shorten the time for making this test. With this end in view, the action of a great number of solvents was ascertained, and it was found that by using 5 cc. of a mixture of equal parts of acetone and carbon tetrachlorid with 5 cc. of the melted fat and allowing this to stand in ice-water, a flocculent precipitate was obtained, when either tallow or hydrogenated oil was present, in from 5 to 30 minutes, depending upon the percentage of adulterant. RECOMMENDATIONS. It is recommended— (1) That the name of this subject be changed from “‘Cocoa and Cocoa Products’ to ‘Cacao Products”. (2) That the present! formule for the polariscopic determination of 1 Assoc. Official Agr. Chemists, Methods, 1916, 329. 498 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 sucrose and lactose be adopted as provisional, and that sucrose and lactose be determined by the polariscopic method', using the present formule. (8) That a 4-hour extraction with anhydrous ethyl ether be adopted as a provisional method for the determination of fat. (4) That the proposed modification of the Baier and Neumann method be further studied with a view to its adoption as a provisional method. (5) That the determination of the critical temperature of dissolution be adopted as a provisional method for the examination of cacao butter. (6) That the associate referee’s test for tallow and hydrogenated oils be adopted as provisional. REPORT ON TEA AND COFFEE. By H. M. Loomis (National Canners Association, Mills Building, Washington, D. C.), Associate Referee. No collaborative work was undertaken. Some preliminary experi- mental work was done, however, in the Bureau of Chemistry, H. A. Lepper and A. W. Broomell collaborating on the determination of mois- ture, sugars and acidity of coffee, and recommendations for modifying the present provisional methods were made to the Committee on Editing Methods of Analysis. With the assistance of Dr. A. Viehoever of the Bureau of Chemistry, a preliminary study of the possibility of basing a standard for raw coffee on the maximum amount of unsound coffee beans and foreign matter was also made. The work on the determination of moisture consisted in the comparison of various methods, but in the absence of any absolute determination of this constituent, the results submitted in Tables 1, 2 and 3 are merely comparative. All of the samples were collected direct from coffee- roasting firms. The work on the moisture content of coffee stored in paper packages in various climates was made primarily to determine whether it would be practicable to set a maximum standard for moisture in roasted coffee in order to prevent the use of excessive water in quenching or coating the coffee. The results in Tables 4 and 5 show the great variations in moisture due to climatic conditions exclusively. As even the most heavily “quenched” or coated coffees seldom contain over 6 per cent of moisture, it is evident that proof of such addition solely by analytical evidence is hardly possible. In this connection the term “‘quenching”’ refers to sprinkling coffee with water in the roasting drums, for the pur- pose of checking the roast, a practice quite common among coffee roasters. 1U.S. Bur. Chem. Bull, 137: 98. 1920] LOOMIS: REPORT ON TEA AND COFFEE 499 “Coated coffee’ refers to coffee which has been treated with sugars, white of egg or some other form of glazing material, for the ostensible purpose of preserving its flavor and aroma. TABLE 1. Green and roasted coffees, ground and analyzed by A. W. Broomell and H. A. Lepper. (December 1914, to January 1915.) DRIED IN VACUUM DRIED IN ATR AT AT 98-99°C. 105°C DESCRIPTION OF COFFEE | Hise B | be nf RESRRE: | Soto: aaa Brazilian: | per cent hours per cent hours ING os eceset Saas 6.22 9 5.80 3) Din netih ii e e eeee 2.43 9 1.97 5 _ Roasted, glazed or coated___________-_-_--- 2.74 9 DIG atl 5 Rio and Victoria blend: DVN. ote eee eee 9.45 6 8.96 s Roasted and quenched __________-------- 1.78 6 1.83 5 Roasted, not quenched _-_-___-_-_------- eT 6 1.33 vi Pocsten ang ieiazed ete te ee 3.24 6 3.22 6 Rio, type No. 7: LI. ou. - be ee re 5.83 | 7 5.85 6 JU Sth eg See a 3.89 | 6 3.98 5 ouarensandiplazed= 2. 2 2). S14 | 6 5.03 5 Victoria, type No. 7: Eee ee bee oe OE a 5.81 6 5.88 5 UINDD RTI Loe eee eee ee 2.99 u. 2.86 6 ovsted annimiazed 222) - 2 8 =. 28 3.98 7 3.96 7 Rio, type No. 4: Rae rec eee. Se ace St 5.76 6 5.93 iT Plnnrerdnste eee = eee cs 3.06 7 3.05 6 Paasted anuupiazed= = 2222-8 es 3.84 7 3.85 7 Victoria: neastediantiplazed= 2 9-2. . =>. Ee | 4.98 7 5.01 ii In the determination of reducing sugars in coffee by the copper reduc- tion method, it was found that the filtration of the cuprous oxid was very materially retarded by a flocculent precipitate, which in some cases entirely stopped filtration. This precipitate was found to contain mag- nesium and iron, probably as hydroxid precipitated by the sodium hydroxid of the Fehling solution. Volumetric methods for the deter- mination of reducing sugars are not applicable because of the deep green color, produced by the alkali, especially with raw coffee, which interferes with the end point. The procedure adopted for this determination was the method for total sugars in foods and feeding stuffs’, with the addi- tion of 1 gram of powdered ammonium sodium phosphate to the sample, 1 Assoc. Official Agr. Chemists, Methods, 1916, 109. 500 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 with the 50 per cent alcohol. This reagent, in the presence of alcohol, precipitates enough of the interfering metals to allow easy filtration of the cuprous oxid, but does not entirely eliminate them, so that the weight of the cuprous oxid can not be used, but the amount of copper reduced must be determined volumetrically or electrolytically. It may be added that this alcoholic digestion method is preferable because raw coffee ferments very rapidly in water solution. Unfortunately, lack of time has made it impossible for H. A. Lepper, who did the work on this method, to submit any figures for reducing sugars obtained by this modified method. TABLE 2. Determination of moisture in one type of Brazilian coffee by different methods. (A. W. Broomell, Analyst.) IN Aner IN IN CURRENT | IN WATER | IN AIR DESCRIPTION OF SAMPLE HEAT- VACU UM VACUUM OF CARBON OVEN AT OVEN AT a at 70°C. At 97°C. | DIOXID AT 97°C. 105°C. 5 98°C. hours per cent per cent per cent per cent per cent PRA WH t= Se Dee ere. eee ee 5 414 5.87 4.86 5.35 5.57 RYN Ee a Sle eee rN i 1.61 6.07 5.56 5.38 5.65 Reatwe 3h oa ie aes 2 9 1.60 6.22 5.76 5.51 5.80 Rawe< 4. aoe 2 3d ioe 11 4.67 6.20 5.77 EIS 5.86 Roasted \1: ane. Lehe See 5 isp 2.10 1.34 1.85 1.97 Roasted :)s) 22a. eee 7 1.76 2.25 1.94 1.82 1.87 Roasted= 45. - eee ae 9 1.69 2.43 2.10 1.74 1.96 Roasted’ 9:2 fey 222 ae ll 1.75 2.25 2.05 1.76 1.97 Roasted and coated________ 5 1.75 2.36 1.70 2.19 2.26 Roasted and coated________ i 2.08 2.62 2.24 2.17 2.15 Roasted and coated________ 9 1.90 2.74 2.39 2.04 2.26 Roasted and coated________ 11 1.95 2.66 2.31 2.05 2.25 TABLE 3. Moisture content of various types of Brazilian coffee. (Dried in air oven at 105°C.) Tuo AND DESCRIPTION OF SAMPLE no No. 4 RIO NO. 7 one yea bypass | per cent per cent per cent per cent per cent Rewer oe Ss ee 11.3 12.0 9.4 11.8 11.0 Roasted before quenching _- ea Oe a — 1.3 1.23 Roasted and quenched ___- 2.1 3.9 2.0 1.6 1.5 After'coating= === 3.7 7.8 3.7 2.7 1.8 1920) LOOMIS: REPORT ON TEA AND COFFEE 501 TABLE 4. Variation in moisture content of coffee stored in 1 pound paper bags. (Dried in air oven at 105°C.) if SORE STORED AND STORED AND | STORED AND | ANALYZED IN seo SEN ANALYZED IN ANALYZED IN | ANALYZED IN DESCRIPTION OF SAMPLE NEW YORK ABOUT 3 % NEW YORK (NEW ORLEANS|NEW ORLEANS | AT TIME OF | aa 8 MONTHS 6 MONTHS | §& MONTHS | COLLECTION AFTER AFTER AFTER | AFTER | COLLECTION COLLECTION COLLECTION | COLLECTION { ! | } Green cofiees: per cent } per cent per cent percent | percent Brazilian blend_________ 11.0 5.86 ae a had oe Rio and Victoria blend __ 11.8 8.88 =e seat | eee TONING: f/2== 55255 2 12.0 5.88 —_ eae) nee RUG ING Sr ae 11.3 5.93 Sat wise ashes WactoriaeNo; «7/22 = --=-- 9.4 5.92 | ree ee pe | Plain roasted: ’ Brazilian blend__-______- 1.0 1.96 4.6 4.83 | 6.24 Rio and Victoria blend_ 1.6 1.86 4.6 pees ped ONG 7 3.9 3.95 = 5.06 6.20 orNos 4 2 £20 be 24 3.09 Sil 4.49 6.34 Waetoria INO! 72—----- = = 2.0 2.89 4.6 4.52 | 6.15 Roasted and coated: Brazilian blend_________ 115) 2.26 4.9 4.79 6.18 Rio and Victoria blend__ Pat 3.29 4.9 Wide ated Li LS Ca (ee 7.8 5.05 oh 6.41 6.43 St Ni ae 3.85 4.9 4.91 6.40 Wictoria No. :72__------- 3.7 3.96 5.0 5.08 6.31 TABLE 5. Variation in moisture content of coffee stored in 1 pound paper bags; samples taken from commercial lots analyzed at different periods after collection. (Dried in air oven at 105°C.) DESCRIPTION OF SAMPLE | 1 montTH | 7 MoNTHS | 8 MONTHS | 9 MONTHS | 10 MONTHS | per cent per cent per cent per cent per cent Roasted and glazed Rio coffee_____ | 8.66 6.46 4.72 5.74 6.62 Rio coffee; low grade. _._-_-_---.- | 5.32 6.21 4.69 5.77 6.50 In connection with a case arising under the Federal Food and Drugs Act, it was desired to determine if the addition of chicory improves in any way the keeping qualities of coffee extract. Several series of hot water extrac- tions of pulverized coffee and coffee-sugar-chicory mixtures were made. Benzoate of soda, and in some instances glycerol, was added, and the number of days before visible mold appeared was recorded. It was found that all extracts of either straight coffee or coffee-sugar-chicory mixture containing benzoate of soda (approximately 0.4 per cent) showed no signs of spoilage in 8 months. Coffee extracts without benzoate of soda spoiled within 9 to 15 days, depending on the concen- tration. Extracts of mixtures of coffee, chicory and sugar containing 502 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 no benzoate spoiled in from 5 to 9 days. Chicory and glycerol (10 per cent) seemed to have no effect upon the keeping quality of the extract. In Table 6 are given analytical results on some samples of coffee, TABLE Analytical resulls on coffees believed (Analyzed by H. A. Lepper and =e mice DESCRIPTION Vere) Or Water- umat| at | Total | insolu-|Soluble| "80l4-| soluble are 98°C. | 105°C. ble . . Brazilian: per cenl| per cent| per cent| per cent| cc.4 ec.4 | percent) per cent Rawisee eee eee ee 6.22¢| 5.86°] 4.49°| 1.31¢) 8.15*} 3.80) 0.196} 0.176 Plain roast, from cooler_____ 2.48e| 1.97¢| 4.26 | 0.79 | 8.75 | 3.25 | 0.202) 0.176 Roasted and glazed ___-_-_- 2.74°| 2.26°| 4.15 | 0.76 | 8.50 | 3.35 | 0.204] 0.167 Rio type No. 7, and Victoria type No. 8: Raye cnc eee eee 9.45 | 9.02 | 4.28¢) 1.09°} 8.10°} 3.70°) 0.177} 0.182 Roasted, not quenched_-__- 1.17 | 1.33] 4.54] 1.05] 8.60] 3.35 | 0.234) 0.189 Roasted and glazed ____-_-- 3.24] 3.29°) 4.27] 0.85 | 8.25] 2.85 | 0.228) 0.172 Rio type No. 7 (large bean): CN ees sects aes ess 5.83¢| 5.85¢] 4.10°] 0.89°| 8.25e) 3.80] 0.183 | 0.202 Roasted and glazed _--_-_--_- 5.14¢| 5.11°] 4.31 | 0.82 | 8.35 | 3.25 | 0.279] 0.165 Plamiroas tess 3.88°| 3.94] 4.31 | 0.94 | 8.60) 2.95 | 0.221) 0.154 Victoria type No. 7: AWein es = ee ee 5.8le| 5.88 | 4.15*] 1.31¢) 7.10°) 3.45°) 0.180} 0.147 Plainwrosst esse es 2.99¢| 2.88] 4.94] 1.86 | 7.60 | 3.15 | 0.145} 0.213 Roasted and glazed _--_-_-_- 3.98e] 3.96) 4.19] 1.14] 7.50} 3.30 | 0.157) 0.185 “Ttalian Roast’? coffee (with- out oil finish) -— ——- es: 4.83 | 4.82°) 4.68 | 0.97 | 9.00] 3.00 | 0.195} 0.194 Victoria, roasted and glazed___| 4.98*} 5.01¢| 4.30 | 0.80] 8.60] 2.60 | 0.206} 0.189 Rio type No. 4: Ch, Rae ee So = ee 5.76¢| 5.92°) 3.8l¢] 0.88¢] 7.55*| 3.55°] 0.144] 0.198 Plamroaste= oe eee ee 3.06¢|} 3.05¢] 4.02 | 0.94 | 7.55 | 3.05 | 0.123} 0.222 Roasted and glazed _____--- 3.85¢| 3.85¢] 4.00 | 0.93 | 7.75 | 3.40 | 0.154] 0.221 ® Nitrogen determined by Mr. T. C. ‘ Trescot, Bureau of Chemistry, Washington, D. C. » Extraction made with Johnson extractor instead of Soxhlet, as directed in Gorter method (U.S. Bur. Chem. Bull. 137: 106). Determined nitrogen direct in 50 ec. of 55 ce. filtrate instead of extracting a second time. © Moisture calculation based on the determination in vacuum at 98°C. * 1920) LOOMIS: REPORT ON TEA AND COFFEE 503 believed to be authentic commercial samples, collected from coffee roasters in New York City during October 1914. In conclusion, it is recommended that the recommendations of the referee for 1915 be carried out, as far as possible, by the referee for 1917. 6. to be authentic commercial samples. A.W. Broomell, spring of 1915.) | CALCULATED TO MOISTURE-FREE BASIS® ABSOLUTE COLD | ETHER pce WATER | CAFFEIN? 4 | Cold NITROGEN® | EXTRACT EXTRACT | Carte: Nitro- Crude Ether 2 Caffein gen fiber eateact water ? i ri extract | | per cent per cent | per cent | per cent | percent | percent | percent | per cent | percent | per cent 2.07 14.35 | 18.39 | 29.82 1.00 1.06 2.20 | 19.60 | 15.30 | 31.79 2.27 14.66 | 19.90 | 22.17 1.14 1.16 2.30 20.36 | 14.99 | 22.68 2.20 14.63 | 20.26 | 21.80 1.16 1.19 2.31 20.81 | 15.02 | 22.39 2.18 12.73 | 16.45'| 28.53 | 0.99 1.09 2.40 14.25 | 17.13 | 22.27 | 1.248) 1.25 z 1.16 2.30 13.79 | 16.62 | 22.95 1.15 17.17 | 14.25 | 23.71 11 12.85 | 14.90 | 29.98 | 1.03 1.09 2.24 15.82 | 13.64 | 31.83 2.18 13.03 | 17.83 | 21.53 | 1.22 1.28 2.29 18.79 | 13.73 | 22.69 2.26 14.05 | 17.26 | 21.62 | 1.31 1.36 2.35 17.95 | 14.61 | 22.49 2.04 14.19 | 17.85 | 27.45 | 0.87 0.92 2.16 18.95 | 15.06 | 29.14 2.19 15.37 | 16.96 | 22.06 | 1.09 1.12 2.25 17.48 | 15.84 | 22.74 2.15 15.00 | 15.77 | 22.19 | 1.14 1.18 2.23 16.40 | 15.60 | 23.08 2.15 14.42 | 14.98 | 22.57 | 0.98 1.03 2.26 15.76 | 15.17 | 23.75 2.18 14.54 | 16.15 | 30.49 1.08 1.14 2.31 17.13 | 15.42 | 32.35 2.35 14.43 | 16.90 | 22.64 | 1.24 1.28 2.42 17.43 | 14.88 | 23.35 1.25 1.30 2.47 2.38 14.25 | 16.11 | 22.59 16.75 | 14.82 | 23.49 4N/10 per 2 gram sample. ae * Analysis made by A. W. Broomell, Bureau of Chemistry, Washington, D. C. u ae {Determination made by G. P. Walton, Cattle Food Laboratory, Bureau of Chemistry, Washington, Gravimetric determination. 504 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 REPORT ON PRESERVATIVES. By A. F. Seeker! (Bureau of Chemistry, Food and Drug Inspection Station, New York, N. Y.), Associate Referee. Following the recommendations adopted at the last meeting, the work for this year has consisted principally of a trial of certain methods for the determination of saccharin. The method given in Bureau of Chem- istry Bulletin 107 (Revised), page 183, is somewhat vague and lacking in details, but its revised form? has been found by the referee to be satisfactory and sufficiently accurate in most cases. It is inaccurate, however, in the presence of ether-soluble sulphur compounds, and in many cases, as pointed out by Gnadinger*, the solution prepared as directed gives rise to troublesome emulsions during extraction. These defects Gnadinger proposes to overcome by treatment of the solution with lead acetate before extraction to precipitate emulsion-forming sub- stances, and the removal of ether-soluble sulphur compounds from the extracted residue by treatment with petroleum ether and later with bromin. The methods selected for trial were substantially those proposed by Gnadinger at the last meeting, the details being as follows: METHOD I. DETERMINATION OF SACCHARIN IN THE ABSENCE OF OTHER ETHER-SOLUBLE SULPHUR COMPOUNDS. PREPARATION OF SOLUTION. Fruit juices, sirups and similar liquid preparations —Transfer 100 grams of the sample to a 250 ce. volumetric flask by means of a little water, dilute to about 200 ce. with water, add 5 cc. of glacial acetic acid, mix, add a slight excess of 20% neutral lead acetate solution, mix thoroughly, dilute to the mark with water, again mix thoroughly and filter through a folded filter. Solid or semi-solid preparations.—Transfer 50 grams of the sample to a 250 ec. yolu- metric flask by means of a little hot water and add sufficient nearly boiling water to make the volume about 200 cc. Allow the mixture to stand for 2 hours, shaking occasionally. Then add 5 cc. of glacial acetic acid, mix thoroughly, add a slight excess of 20% neutral lead acetate solution, dilute to the mark with cold water, mix, and allow to stand for 20 minutes. Filter through a folded filter. EXTRACTION AND DETERMINATION. Transfer 150 cc. of the filtrate to a separatory funnel, add 15 ec. of concentrated hydrochloric acid and extract three times with 80 cc. portions of ether, shaking the separatory for 2 minutes each time. Wash the combined ether extracts once with 5 ce. of water, remoye the ether by distillation and transfer the residue to a platinum crucible by means of a little ether, or if substances diflicultly soluble in ether are present, use alternate small portions of water and ether. Evaporate the ether on a steam bath, ' Since deceased. * Assoc. Official Agr. Chemisls, Methods, 1916, 145. * J. Assoc. Offictal Agr. Chemists, 1917, 3: 25. 1920) SEEKER: REPORT ON PRESERVATIVES 505 add 2-3 cc. of a 10% sodium carbonate solution to the residue, rotate so that all of the residue is brought into contact with the solution and evaporate to dryness on a steam bath. To the dry residue in the crucible add 4 grams of a mixture of equal parts of anhydrous sodium and potassium carbonates, heat gently at first and then to complete fusion for 30 minutes over an alcohol or other sulphur-free flame. The fusion may be conducted with a gas flame by closely fitting the crucible into a hole cut into a piece of heavy asbestos board so that one-third of the crucible projects above the asbestos, and heating the lower portion of the crucible by means of a large Bunsen or Meker burner. Cool, dissolve the melt in water, add about 5 cc. of bromin water, acidify with hydrochloric acid, filter, wash the paper with a little water, dilute the filtrate and washings to about 200 cc., heat to boiling and slowly add an excess of barium chlorid solution. Allow the mixture to stand 5-6 hours, or better, overnight, separate the precipitated barium sulphate by filtration, wash until free from chlorids, dry, ignite, cool and weigh. Correct the result thus obtained for any sulphur present in the fusion mixture as found by a blank determination. Calculate the equivalent amount of saccharin by multiplying the corrected weight of barium sulphate by 0.7845. Note.—Instead of the mixed sodium and potassium carbonates, 3-4 grams of sodium peroxid may be employed for the fusion. In this case, a nickel crucible must be used and the time of fusion may be reduced to 5 minutes. The separation of a little lead chlorid during the extractions does not interfere with the accuracy of the method. METHOD II. DETERMINATION OF SACCHARIN IN THE PRESENCE OF MUSTARD OIL. PREPARATION OF SOLUTION. Prepare the solution as directed under “‘Method I, solid or semi-solid preparations’. EXTRACTION AND DETERMINATION. Transfer 150 cc. of the filtrate, obtained as directed under ‘‘Preparation of Solution’’, to a separatory funnel, add 15 cc. of concentrated hydrochloric acid and extract three times with 80 cc. portions of ether, shaking the separatory for 2 minutes each time. Wash the combined ether extracts once with 5 cc. of water and transfer the ether to a 250 cc. beaker. Add about 10 grams of washed sea sand and evaporate the ether before a fan or air blast. Distribute the sand on the walls of the beaker with a stirring rod and continue the spontaneous evaporation until quite dry. Add 25 cc. of petroleum ether (b. p. 30-65°C.), and rub thoroughly with a “policeman”. Decant through a dry 7 cm. quantitative paper and repeat the washing twice, using 25 cc. of petroleum ether each time. Reject the petroleum ether washings and return the filter paper to the beaker containing the sand. Wash the residue on the sand with hot water and filter into a separatory funnel, collecting about 75 cc. of the filtrate. Cool, add 7-8 cc. of concentrated hydrochloric acid and a distinct excess of bromin water, allow to stand for 5 minutes and destroy the excess of bromin with sodium nitrite solution, avoiding a large excess of the latter. Extract the acid solution three times with 50 cc. portions of ether, and wash the combined ether extracts once with 5 ce. of water. Eyaporate the ether, subject the residue to alkaline fusion, and precipitate barium sulphate in an acidified solution of the melt, as directed under Method I, beginning with “Cool. dissolve the melt in water, etc.” Multiply the corrected weight of barium sulphate found by 0.7845 to obtain the equivalent weight of saccharin, and add 0.5 mg. to this result to correct for the saccharin dissolved by the petroleum ether. A preliminary trial of these methods by the referee gave the results shown in Table 1. 506 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TABLE 1. Resulls of delerminations of saccharin by Methods I and II. | SACCHARIN SACCHARIN | SUBSTANCE METHOD ADDED FOUND RECOV ERY | | per cent per cent per cent Raspbexcvigon! 426 > eee ee 4 i | 0.050 0.048 96 Sweetpickless. st = eee ore ; ] | 0:075 0.069 92 Cider! 4 eae) eet 2 ios oh aie arerbod h ti £0:020 0.019 95 Ketchup tae ofgewi ke su £ ve Ae Rh I | 0.100 6.092 92 Rietich tpi tte ae te hk ee ai | I | 0.050 0.049 98 Chow chow. -- = ----- , Ne Je et. oR If 0.050 0.043 86 Chow CuOWw. ooo a2 eee eee If | 0.050 0.046 92 Ghow chow! -DEstria! Oper ci qa Eee | O00 0.082 82 Ghow: chow: fieiblwr sack ack Bye el 1 | 0.100 0.088 88 The results given in Table 1 show an average recovery by Method I of 95 per cent and of 87 per cent by Method II. The results obtained by similar methods as reported by Gnadinger at the last meeting were: Average recovery, Method I, 99 per cent; and Method II, 92 per cent. The success of the preliminary work warranted submitting these methods to further trial by the collaborators. Accordingly, three mixtures were prepared and sent to the various collaborators, together with the details of the methods as given aboye. The substances selected for the test were of such a character as to represent as nearly as possible the different types in which saccharin is commonly found, and which would present the ordinary difficulties of accurate determination, but which would, at the same time, allow the saccharin to be uniformly distributed throughout the sample so that the portion taken by each analyst for the determination would contain the same amount of the substance sought. The first mixture consisted of strawberry sirup prepared by the referee from strawberry juice and cane sugar to which 0.066 per cent of saccharin was added. The second mixture consisted of a commercial tomato puree, in which no saccharin was found by a blank determina- tion, to which 0.066 per cent of saccharin was added by the referee to- gether with 0.5 per cent of boric acid to prevent decomposition before analysis. The third mixture consisted of a finely ground mustard relish prepared by the referee from cucumbers, onions, cauliflower, red pepper, curry powder, salt and vinegar, 0.099 per cent of saccharin being added to the finished product. The results obtained by the ten collaborators as given in Table 2 were reported by: (1) S. Adler, U. S. Bureau of Animal Industry. Fed- eral Building, Kansas City, Kans.; (2) C. B. Gnadinger, U. S. Food and Drug Inspection Station, Transportation Building, Chicago, IIl.; (8) M. E. Hinds, Food and Drug Department, Nashville, Tenn.; (4) L. Katz, U. S. Food and Drug Inspection Station, U. S. Appraiser’s 1920] SEEKER: REPORT ON PRESERVATIVES 507 Stores, New York, N. Y.; (5) H. B. Mead, U. 8. Food and Drug In- spection Station, U.S. Appraiser’s Stores, Philadelphia, Pa.; (6) L. C. Mitchell, U. S. Food and Drug Inspection Station, Old Custom House, St. Louis, Mo.; (7) M. B. Porch, H. J. Heinz Company, Pittsburgh, Pa.; (8) W. D. Richardson, Swift & Company, Chemical Laboratory, Chicago, Ill.; (9) M. G. Wolf, U. S. Food and Drug Inspection Station, U. 8. Appraiser’s Stores, New York, N. Y.; and (10) P. B. Yost, U. S. Food and Drug Inspection Station, U.S. Custom House, New Orleans, La. To all of these collaborators the referee wishes to express his acknowledgments for their valuable cooperation. TABLE 2. Results obtained by collaborators on samples submitted by referee. Mernop I Mertnop II STRAWBERRY SIRUP TOMATO PUREE MUSTARD RELISH (CONTAINED 0.066 PER CENT | (CONTAINED 0.066 PER CENT | (CONTAINED 0.099 PER CENT ANALYST OF SACCHARIN) OF SACCHARIN) OF SACCHARIN) Sete Recovery pmpuay Recovery frecuat Recovery per cent per cent per cent per cent per cent per cent (1) 0.070 106 0.066 100 0.076 0.068 103 0.072 109 0.072 73 0.049 74 0.067 102 0.073 74 (2) 0.052 79 0.062 94 0.083 84 0.058" 88 0.065 99 0.088 89 (3) no sa/mple no sa/mple 0.070 mee 0.070 71 (4) 0.062 94 0.065 99 0.088 89 0.050 76 0.062 94 0.088 89 see bh =a 's= ise 0.084 85 (5) 0.062 94 0.065 99 0.070 ti 0.065 99 oo se 0.075 76 (6) 9.059 89 0.048 73 0.070 OL 0.060 91 0.053 80 0.052 53 (7) 0.063 96 0.042 64 no salmple 0.065 99 0.050 76 (8) 0.049% 74 0.062 94 0.076 77 0.050 76 0.056 85 0.072 73 (9) 0.058 88 0.064 97 0.092 93 0.055 83 0.059 89 ee Be (10) 0.066 100 0.067 102 0.083 84 0.067 102 0.069 105 0.080 81 * Sample badly fermented. 508 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 COMMENT BY COLLABORATORS. C. B. Gnadinger.—The results obtained on Sample 1 (strawberry sirup) were not satisfactory. From the experience I have had with the method, I am led to believe that this is due to the manner of preparing the extracted saccharin for fusion. When ether is evaporated from a crucible on a steam bath, as directed in the method, there will be some loss by spattering; moreover, the contents of the crucible have a tendency to creep over the edge and to form a ring at the top of the crucible. It is difficult to prevent loss of this material. During the extraction a large amount of acetic acid is taken up by the ether and remains (in Method I) after the ether is distilled. It was found that 3 cc. of 10% sodium hydroxid were not sufficient to neutralize this acid. If the acid solution is evaporated to dryness, saccharin may be lost; if sufficient alkali is added to neutralize the acid, the evaporation requires a long time. In the case of Sample 2 (tomato puree), sufficient sodium carbonate was added to make the solution alkaline before evaporating. In every case it was necessary to filter the solution of the melt, even when platinum crucibles were used. The procedure given below is that used in developing the method as reported at the last meeting of the association. ‘his procedure should have been given as part of the method, but was omitted through the fault of the writer. Other analysts and I have examined a number of samples with excellent results. Proceed_as in Method I to “Wash the combined ether extracts once with 5 cc. of water’. Transfer the ether to a beaker and evaporate to dryness before a fan or air blast. Dissolve the residue in about 5 cc. of alcohol, add phenolphthalein and about 5 cc. excess of approximately N/10 sodium hydroxid. Transfer to a 40 cc. nickel crucible with hot water, rinsing the beaker three times. Evaporate on a steam bath until only 1-2 cc. remain. Weigh out 5 grams of fusion mixture (6 parts of sodium carbonate and 1 part of potassium nitrate), carefully rotate the crucible so that the entire inner surface is moistened, and with a spatula add part of the fusion mixture, completely covering the moistened wall of the crucible. Finally scrape the material from the wall to the bottom of the crucible and cover with the rest of the fusion mix- ture. Cover the crucible and heat gently for about 5 minutes and then fuse for 25 minutes. Cool, place the crucible in a 250 cc. beaker, cover with cold water and add 15 cc. of concentrated hydrochloric acid. Let stand until the melt is dissolved, filter through a quantitative paper, neutralize with ammonium hydroxid and add 1 cc. excess of concentrated hydrochloric acid. Heat to boiling and add slowly with constant stirring an excess of boiling barium chlorid solution; continue the boiling several minutes and let stand overnight. Collect the precipitate on a weighed Gooch crucible and wash thoroughly with boiling water. Dry, ignite, cool and weigh. Correct the result thus obtained for any sulphur present in the fusion mixture, as found by a blank determination. Calculate the equivalent amount of saccharin by multiplying the cor- rected weight of barium sulphate by 0.7844. The three samples submitted by the associate referee were examined by this procedure and the results are given herewith: STRAWBERRY SIRUP® TOMATO PUREE | MUSTARD RELISH per cent per cent per cent 0.066 0.082 0.092 0.067 0.075 0.088 0.068 0.077 0.091 * Sample badly fermented. L. Katz.—Shaking for 30 seconds instead of 2 minutes was tried and found to give somewhat lower results: Strawberry sirup, 0.047 per cent; and tomato puree 0.060, per cent. 1920] SEEKER: REPORT ON PRESERVATIVES 509 H. B. Mead.—The method, while somewhat tedious, proceeds smoothly. From my experience in using immiscible solvents, I believe it is unnecessary to shake the extrac- tions for 2 minutes. Accordingly all extractions in the first determination in each case were shaken for 2 minutes. The first extraction in the second determination in each case was shaken for 2 minutes, the second and third extractions for 30 seconds each. L. C. Mitchell.—An electric muffle was used in fusing the sodium and potassium carbonates. The platinum crucibles were put into a cold muffle, the current turned on, the temperature raised gradually to complete fusion of the carbonates, and kept at this temperature for 30 minutes. The following method for the determination of sac- charin was tried: Transfer 100 grams of the sample to a 250 cc. flask by means of a little water, dilute to about 200 cc. with water, make distinctly alkaline to litmus paper with strong sodium hydroxid solution, mix thoroughly, dilute to the mark with water, again mix thoroughly, let stand at least 2 hours and filter through a folded filter. Transfer 150 cc. of the filtrate to a separatory funnel, add 15 cc. of concentrated hydrochloric acid and extract with 100, 50, 50, and 50 cc. portions of a 1 to 1 mixture of ether and petroleum ether (b. p. 30-60°C.). Shake carefully for 2 minutes each time. Wash the combined ether petroleum ether extracts once with 5 cc. of water, and allow the ether to evaporate spontaneously. Add 25 cc. of neutral 95% alcohol and titrate with potion hydroxid (1 cc. equivalent to 1 mg. of saccharin), using phenolphthalein as indicator. The following results were obtained: Strawberry sirup, 0.048 per cent; tomato puree, 0.045 per cent; and mustard relish, 0.090 per cent. The titrated liquids were in each case then subjected to the fusion procedure. Sam- ples 1 and 2 as follows: After titrating add 5 cc. of a 10% sodium carbonate solution, evaporate nearly to dryness on a steam bath, transfer to a platinum crucible by means of a little hot water, and evaporate to dryness on a steam bath. Then continue as in Method I. The results obtained were: Strawberry sirup, 0.044 per cent; tomato puree, 0.039 per cent. Sample 3 (mustard relish) was treated as follows: After titrating add 10 grams of washed sea sand, evaporate to dryness on a steam bath, distributing the sand on the sides of the evaporating dish with a stirring rod before complete dryness is reached. Add 25 cc. of petroleum ether and continue as in Method If. The result obtained was 0.077 per cent. Duplicate determinations were not made, owing to the lack of sufficient sample. M. G. Wolf—Two methods in addition to those submitted by the referee were tried. One of these is based upon the determination of saccharin by means of its nitrogen content and in the other the saccharin is weighed as such. The first method is as follows: Extract the saccharin as directed in Method I, proceeding to the point at which the impure saccharin is recovered from the ether solution, as indicated by the words, “Wash the combined ether extracts once with 5 cc. portions of water, remove the ether by distillation”. Transfer the residue to a 50 cc. beaker by means of a little ether and evaporate the ether on a steam bath. Add 10 cc. of water to the residue in the beaker, heat the covered beaker on a steam bath for 5-10 minutes, add 0.05- 0.10 gram of sodium acetate and, when this has dissolved, cool and add 10cc. of alcohol and 0.5 cc. of a saturated aqueous solution of silver nitrate. Allow the mixture to stand overnight, collect the precipitate on a Gooch crucible and wash with 40-50 cc. of alcohol in small portions at a time. Dry the crucible in a water oven, and then transfer the asbestos pad with the precipitate to an Erlenmeyer flask. Add sufficient water to the flask to make the volume of liquid about 100 cc., then add 10 cc. of con- centrated hydrochloric acid for each 90 cc. of liquid, insert a short-stemmed funnel into the neck of the flask and boil gently for 45 minutes. Filter through a Gooch and wash with hot water, collecting the filtrate and washings in a porcelain dish. To the combined filtrate and washings, add an amount of platinic chlorid solution equivalent to 0.2-0.3 gram of PtCl, (in case a large amount of extractive matter remains after 510 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS |[Vol. IIT, No. 4 removing the ether, use 0.5 gram) and evaporate on a steam bath to a pasty consistency. Then add about 25 cc. of 80% alcohol, collect the precipitate of ammonium platinic chlorid on a Gooch, wash with 80% alcohol, dry at 100°C., cool and weigh. Wash the residue on the Gooch repeatedly with hot water to remove the ammonium platinic chlorid, finally wash with a little 95% alcohol, dry at 100°C., cool and weigh. The difference between the first and second weighings is the weight of the ammonium platinic chlorid obtained from the decomposition of the saccharin. Multiply this weight by 0.825 to obtain the equivalent of saccharin. Method in which the saccharin is weighed as such: Proceed as in the foregoing method to the point indicated by the words, “‘Allow the mixture to stand overnight, collect the precipitate on a Gooch crucible and wash with 40-50 ce. of alcohol in small portions at a time’. After completing the operation just mentioned, wash with three small portions of ether, dry in a water oven, and transfer the asbestos pad and the precipitate to a beaker. Add 50 ce. of water and sufficient bromin water to produce a strong coloration, and heat on a steam bath with frequent stirring for about 10 minutes. Filter through a Gooch crucible, and wash with successive small portions of hot water until the volume of the filtrate measures 100-120 cc. Cool the combined filtrate and washings, transfer to a separatory funnel, add sufficient sodium bisulphite to destroy the excess of bromin, add 5 ce. of concen- trated hydrochloric acid for each 100 cc. of liquid, and extract four times with washed ether, using for each extraction a volume of ether equivalent to half the volume of the aqueous layer. Wash the four ether extracts in succession with two 5 ce. portions of water, transfer the ether solution to a flask, distil off the solvent, then transfer the residue by means of a little ether to a small tared dish, evaporate the ether on a steam bath, and dry the residue of saccharin to constant weight at 100°C. The result obtained by the first method on the tomato puree was 0.057 per cent of saccharin, and by the second method on the strawberry sirup, 0.064 per cent of saccharin. DISCUSSION OF RESULTS. Taking into account the small amount of saccharin present in the samples, which would tend to magnify ordinary errors when results are compared upon the basis of percentage yield, it is considered that a recovery of saccharin coming within 15 per cent of the actual amount present may be regarded as within the permissible limit of error. Upon this basis and by disregarding those results in which the strawberry sirup was badly fermented when analyzed, it is found that, in the case of the strawberry sirup, twelve of the fifteen results reported and, in the case of the tomato puree, fourteen of the eighteen results reported, may be accepted as satisfactory, all of these having been obtained by Method I as submitted by the referee. The majority of these results (twenty-two out of thirty-three) come well within a 10 per cent limit of error. The method may therefore be accepted as satisfactory. With regard to Method IT, only five of the nineteen results reported come within the 15 per cent limit of error. The results are all low, and with the exception of one, which may be disregarded, show a yield of 71 to 93 per cent of the actual amount present. It appears that some source of error exists, and further work should be done upon this method to ascertain whether the details suggested by C. B. Gnadinger in this report or other modifications will correct its defects. It is desirable that a method for the determination of saccharin other than one based upon its sulphur component be found for purposes of 1920} SEEKER: REPORT ON PRESERVATIVES 511 confirmation, in order that in contested cases the findings may be placed upon an unassailable basis. Owing to the great amount of time de- manded by the associate referee’s duties as a member of the Committee on Editing Methods of Analysis, it was found impossible to continue the work done last year along this line. For this purpose, methods such as those proposed by M. G. Wolf in this report should be investigated. Another feature of the work still remaining unsolved is the deter- mination of saccharin in baked flour preparations of the nature of ice cream cones, the difficulties of which were pointed out last year. RECOMMENDATIONS. It is reeommended— (1) That Method [, as given in this report, for the determination of saccharin in the absence of other ether-soluble sulphur compounds be adopted as a tentative method in place of the existing provisional method. (2) That further work be done on Method II for the determination of saccharin in the presence of mustard oil, that other methods not dependent upon the sulphur component of saccharin be investigated, and that further work be done upon the determination of saccharin in baked flour preparations. SALICYLIC ACID. The attention of the associate referee has several times been called to the fact that only one qualitative test for salicylic acid is given in Bureau of Chemistry Bulletin 107 (Revised), and that this in some instances may lead to erroneous findings. It has been the experience of the referee that the ferric chlorid test there given is not characteristic for salicylic acid, and that a similar reaction is given by the residue from the ether extract of roasted or caramelized cereal products. Others have observed the same fact, as shown by reference to the literature: Abraham!; Erich?; Brand’; Munsche*; Backe®; Sherman®; Sherman and Gross’. The reaction proposed by Jorissen® has been in use in the New York Food and Drug Inspection Station of the Bureau of Chemistry for several years, and has proved very satisfactory as a confirmatory test. The method was first published in 1882 and has been favorably reported at various times since then: Klett®; Windisch'®; da Silva'!; Portes and 1 Abst. Z. Nahr. Genussm,, 1898, 1: 857. ; Bierbauer, 1893, 24: 465. 3 Z. ges. Brauw., 1893, 16: 303; Ber., 1894, 27: 806. 4 Wochschr. Brauerei, 1893, 10: 739. 5 Compl. rend., 1910, 150: 540; 1910, 151: 78. ® J. Ind. Eng. "Chem., 1910, 2: 24. 7 Thid., 1911, 3: 492. 8 Bull. acad. roy. Belg., 1882, 3rd ser., 3: 259. ® Pharm. Centrh., 1900, 41: 452. 10 Z. Nahr. Genussm., 1903, 6: 447. 11 Compt. rend., 1900, 131: 423. 512 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 Desmouliéres!; Sherman?; Sherman and Gross’. C. L. Black of the Philadelphia Food and Drug Inspection Station of the Bureau of Chem- istry, also reports favorably on the test. It has been stated by several of those who have used the test that the reaction is distinct when as little as 0.05 mg. of salicylic acid is present in 10 cc. of the solution tested. This has been confirmed by the referee, who also finds that, with the exception of coloring matter, none of the substances ordinarily present in the ether extracts of foods interfere with the test, and no interfering substance is to be expected if the salicylic acid is purified in the usual way. Special interest is to be attached to the fact that neither benzoic nor cinnamic acids give the test. The reaction is performed as follows: Extract the salicylic acid as directed in the tentative method‘. Dissolve the purified salicylic acid in hot water, cool 10 cc. of the solution in a test tube, add 4 or 5 drops of a 10% potassium nitrite solution, 4 or 5 drops of 50% acetic acid and 1 drop of 10% cupric sulphate solution, mix thoroughly and heat to boiling. Boil for 30 seconds and allow to stand for 1-2 minutes. In the presence of salicylic acid a blood red color will develop. Taking into consideration the length of time this test has been in use, the numerous favorable reports in the literature and the experience reported by the New York and Philadelphia Food and Drug Inspection Stations of the Bureau of Chemistry, together with the need of an offi- cial confirmatory test to supplement the findings by the ferric chlorid reaction, which alone may prove misleading, it is recommended that the Jorissen test as here given be tentatively adopted. Messrs. H. D. Gibbs® and G. A. Geiger® (Bureau of Chemistry, Wash- ington, D. C.) presented a paper on “The Manufacture of Benzaldehyde with Benzoic Acid as a By-Product’”’. REPORT ON METALS IN FOOD. By Daviw Kier’ (Division of Foods and Dairies, Illinois Department of Agriculture, 1410 Kimball Building, Chicago, IIl.), Associate Referee. The work was confined to the study of the Gutzeit method for arsenic, and its application to specific foods. It was also intended to study the methods for the determination of tin. This part of the work was un- fortunately interrupted when the chemist who was assigned the problem 1 Ann. chim. anal., 1901, 6: 401. 2J. Ind. Eng. Chem., 1910,)2: 24. 2 Tbid., 1911, 3: 492. * Assoc. Official Agr. Chemists, Methods, 1916, 141. 6 Present address, Jackson Laboratory, E. I. Du Pont a Wilmington, Del. 6 Present address, 43 Gaston Street, West Orange, N. 7 Present address, Hollister-Wilson Laboratories, chon go, Il. — 1920) KLEIN: REPORT ON METALS IN FOODS 513 was called out with the troops. The work has recently been resumed, but nothing can be reported at this time. In view of the very unsatisfactory and conflicting results reported by previous collaborators on the Gutzeit method, it was decided not to send out any samples until the method was subjected to close scrutiny, and the doubtful directions revised. To that end Mr. J. J. Doyle (Illinois Division of Foods and Dairies, Chicago, Ill.) has devoted most of his time during the past nine months. Certain changes in the previous pro- cedure can be reported now, which should lead to more accurate and uniform results in the hands of different chemists. One of the changes is the use of Munktell’s No. 00 filter paper in place of Whatman’s cold pressed paper for making the stains. With the latter paper it was not possible to obtain stains of equal density or length on both sides of the paper. The results were erratic and unsatisfactory. ‘Munktell’s No. 00 paper was selected after trying several kinds of hard and soft papers. With this paper it is not difficult to obtain easily reproducible stains of equal density and size on both sides of the paper. The width of the test paper should be that of the diameter of the tube, rather than slightly smaller. With a paper narrower than the tube, there was a tendency to curl, causing a deflection of the gas current and an unevenness of deposit. Great care should be taken in getting tubes of the same diameter. In our work, the tubes had a diameter of } inch. The best concentration of sensitizing solution was found to be 1.5 per cent mercuric bromid solution. In preparing the test paper, blotting the excess liquid was found unsatisfactory. Better results were obtained by allowing the papers to dry in the air. For generating the hydrogen and the arsine, sulphuric and hydro- chloric acid yield identical results. The concentration of the acid, if sulphuric is used, should be 3 to 4 cc. of acid (sp. gr. 1.84) in 40 cc. of total liquid; if hydrochloric is used, 4 to 8 cc. of acid (sp. gr. 1.20) in 40 ec. of total liquid. The presence of tin and iron in the reacting liquid was thoroughly investigated. Stannous compounds are essential. When stannous chlorid was omitted, not only was the evolution of arsine slow, but it was also very incomplete. Since the injurious effect of ferric iron has been well established, no work was done on it. The literature on the influence of ferrous iron is conflicting. Our experiments indicate that the presence of ferrous iron hastens the complete evolution of the arsine, but is with- out influence on the final amount evolved. For this reason a small amount of ferrous iron is added, which is further treated with stannous chlorid to insure its complete existence in that state. The presence of potassium iodid offers no advantage in the procedure. Its use was discontinued. 514 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 In the construction of the apparatus, the coil of filter paper was replaced by cotton moistened with a 20 per cent lead acetate solution. The temperature of the reacting solution should be more accurately controlled than was formerly the case. Immersion of the bottles in water maintained at about 25°C. throughout the reaction was found to be a convenient method of controlling the temperature and securing a uniform evolution of arsine. A method was devised for estimating the amount of arsenic in an unknown stain without a direct comparison with standard stains. Inas- much as the standards do not keep well and are troublesome to handle, the suggested method simplifies the determination without any loss of accuracy. Differences of 0.5 microgram are readily detected. Advan- tage is taken of the fact that the upper limits of the stains for correspond- ingly increasing amounts of arsenic lie upon a straight line. If, therefore, the slope of this line is determined and the line drawn to correct scale, the amount of arsenic on any test paper can be quickly ascertained by super- imposing it on the scale at the point where the upper limit of stain coin- cides with the straight line. In place of using the extreme limit of colora- tion as the basis of comparison, a point slightly below this has been chosen arbitrarily, where the dark brown abruptly changes into the pale yellow. As applied to foods, the method was thoroughly tested on a phosphate starch mixture used as part of the ingredients in baking powder. Despite the disadvantages and inconvenience of a wet digestion method, we have not found a method superior to it. Work along this line is still in progress, but all results herein reported were made after destroying the organic matter with nitric and sulphuric acid. The often quoted precau- tion of avoiding high temperatures during the digestion was subjected to considerable study. It was found that the loss of arsenic through volatilization, if it occurred at all, was too slight to be detected. This is also true, even if the product contains a large amount of chlorid. For these reasons no special precautions were deemed necessary in the main- taining of definite temperatures during the digestion. METHOD FOR THE DETERMINATION OF ARSENIC. (Adapted from method of C, R. Smith*.) REAGENTS. Concentrated nitric acid and concentrated sulphuric acid—Must be arsenic-free. Zinc.—Arsenic-free. stick zinc broken in pieces to weigh about 10 grams. The zine may be used again if its surface remains crystalline. After a test, wash the zine in distilled water, let it dry on filter paper. 1U.S. Bar. Chem. Circ. 102; J. Soc. Chem. Ind., 1907, 26: 1115; Original Contributions, Eighth Intern. Cong. Appl. Chem., 1912, L: 9. 1920] KLEIN: REPORT ON METALS IN FOODS 515 Lead acetate colton.—Absorbent cotton, part of which is soaked in 5% lead acetate solution, and part in 20% solution. Squeeze out excess moisture. Mercurie bromid paper.—Immerse 15 cm. Munktell’s No. 00 filter paper in the solu- tion contained in a tall, narrow beaker. Allow the paper to remain a few minutes. Slowly draw it out of the liquid, touching the paper to the side of the beaker. Place the paper on a flat surface so that the paper forms a convex arch. Allow it to dry thoroughly. With a photo-trimmer or other suitable device, cut the test papers to fit the tubes as accurately as possible. Reject all outside portions of the original paper. Before placing the paper in the tube, cut off about 1 cm. from one end of the paper. The freshly cut edge should form the base of the test paper. The 1 cm. strip can be used as indicated in the directions. Stannous chlorid solution—Forty grams of the crystals made up to 100 cc. with con- centrated hydrochloric acid and water. Standard arsenic solution.—Dissolye 1 gram of arsenious oxid in 25 ce. of 20% sodium hydroxid, neutralize with dilute sulphuric acid. Add 10 cc. of concentrated sulphuric acid and dilute to 1 liter with recently boiled distilled water. 1 ce. of this solution contains 1 mg. As.Q03. Dilute 20 ce. of this solution to 1 liter. Make 50 cc. of the dilute solution up to 1 liter. Each ce. of the latter solution contains 0.001 mg. of arsenic trioxid. This solution is used to make the standards. The dilute solutions should be made up freshly when required. Ferrous ammonium sulphate—Dissolve 30 grams of the crystals in water, slightly acidified with sulphuric acid; then dilute to 100 cc. APPARATUS. The generator is a 2 ounce wide-mouthed bottle. This is connected by means of a rubber stopper to a glass tube, 1 cm. 6 cm., loosely filled with the absorbent cotton soaked in 20% lead acetate solution. It is advisable not to pack the lower end of the tube, thus preventing the liquid carried up by the gas bubbles from acting on the lead acetate. This lower tube is connected with a similar one, loosely packed with cotton soaked in 5% lead acetate solution. The cotton should be uniformly moist in all tubes. The second tube is connected with a capillary tube exactly 3 mm. in internal diameter and about 12 cm. in length. It is advisable to smooth the upper end of this tube by filing rather than fire-polishing it, unless the latter is very carefully done, so as to prevent reduction of the diameter. The sensitized mercuric bromid paper is carefully placed in this upper tube with the freshly cut end downward. All connec- tions are made with rubber stoppers which should be free from any white coating. PREPARATION OF THE STANDARDS. Into each generator bottle put about 40 cc. of sulphuric acid solution containing 3.5 cc. of concentrated acid. Add 0.5 cc. each of the stannous chlorid and ferrous ammonium sulphate solutions. Add the requisite amount of the standard arsenic solution to each generator bottle, so that the stains will represent 0.5, 1, 1.5, 2, 3, 4, 6, 8, 10 micrograms of As.O;. Keep the bottles in a water bath at 90°C. for about 10 minutes. Test a drop of the solution for ferric iron with potassium sulphocyanate. If no ferric iron is present, place the bottles in a water bath at 25°C. Allow sufficient time for the bottles to attain that temperature. Add 2 pieces of stick zine about 1 inch long. Immediately attach the rest of the generating apparatus, into the upper tube of which the test paper has been fitted. Maintain the water in the bath at 516 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 25°C., and allow the reaction to proceed for about 45 minutes. At the end of that time, remove the test papers, and put in short lengths of fresh papers, to make sure that the evolution is complete. Allow the papers to dry in the air for a few minutes; then immerse them for a moment in melted paraffin. On properly ruled cross section paper, with the width of the strips as the abscissa, place the strips in serial order, leaving the necessary intervals for the concentrations lacking to make a complete series separated by 0.5 microgram. If the series is properly prepared, the upper edge of the stains should lie upon a straight line. Instead of using the extreme upper color limit, it has been found simpler to select a region slightly below this, where the darker color seems to change abruptly into a pale yellow. This point on the 0.5 and 10 microgram stains should be marked on the cross section paper. Through these points draw a straight line. Draw vertical lines to intersect the slanting line, at 0.5 microgram intervals. The amount of arsenic on any unknown strip may be read directly from this scale by placing the strip in the proper place, so that the upper edge of the dark coloration will coincide with the scale. After the scale has been established, the standard stains need not be retained. It is not advisable to work with quantities giving stains of more than 10 micrograms. DETERMINATION OF ARSENIC IN BAKING POWDER AND BAKING POWDER CHEMICALS CONTAINING ORGANIC MATTER. Weigh 10 grams of powder into a dry 4 inch porcelain casserole. Add 10 ce. of nitric acid (sp. gr. 1.42). Cover the casserole with a watch glass and warm gently on a steam bath until the starch is hydrolyzed and nitration begins. Remove from the bath, and set aside until the violent reaction is over. Then add 8 cc. of sulphuric acid (sp. gr. 1.84) and heat on a steam bath for about 30 minutes. Continue the heating on a hot plate until the liquid turns dark brown, but avoid an excessive charring. Remove from the hot plate, add 2 cc. of nitric acid and heat again until the appearance of charring. Repeat the addition of nitric acid and subsequent heating until there is no sign of charring after the nitric acid is driven off. There should be a slightly yellow solution with a mass of calcium sulphate. When the destruction of organic matter is complete, remove the casserole from the hot plate, wash down the contents with 50-60 cc. of water, and evaporate first on a steam bath, then on a hot plate until fumes of sulphur trioxid appear. Remove from the bath and repeat the operation until all of the nitric acid is driven off. This may be determined by adding a drop of the liquid to a drop of diphenylamin sulphate solution (40-50 mg. in 2 cc. of concentrated sulphuric acid) on a porcelain plate. If nitric acid is present the mixture will turn blue. This test is extremely delicate, a blue color often resulting from air contamination. Transfer the contents of the casserole to a 100 cc. volumetric flask. Make up to the mark; let the precipitate settle and pipette 50 cc. of the supernatant liquid into a Gutzeit generator bottle. Add 1.5-2 cc. of stannous chlorid solution and proceed as directed when making the standards. It may happen that the amount of stannous chlorid is not sufficient to reduce the iron. In that case add 1 cc. of stannous chlorid and repeat until all iron is reduced. In the presence of large amounts of phosphates and sulphates, the potassium thiocyanate is seldom colored red by traces of ferric iron. A light brown or straw color usually results. Estimate the amount of arsenic on the test paper by superimposing it on the proper place of the scale. Paraffining the test paper aids in its keeping quality, especially if it is kept in a cool, dark place. * se 1920] KLEIN: REPORT ON METALS IN FOODS 517 The following data indicate the results that can be expected, the degree of concordance and accuracy. TABLE 1. Results of experiments on a commercial phosphate-starch mizture to which varying amounts of arsenic triorid were added before digestion. | AMOUNT DIGESTED | JoTAL ABSENIC ADDED ARSENIC AMOUNT ADDED SAMPLE DIGESTED Osrsne a Starts Se SRO MID ONITEEE pe A grams grams parts per million parts per million parts per million 5.0 2.5 0.4 0.0 a 10.0 5.0 « 0.8 | 0.0 =e 10.0 5.0 0.7 0.2 0.1 10.0 5.0 1.6 1.0 1.0 10.0 5.0 2.0 2.0 1.4 10.0 5.0 0.65 0.0 ae 10.0 5.0 | 0.7 0.2 0.1 1 10.0 5.0 1.8 | 1.0 1.2 10.0 5.0 1.6 2.0 1.0 10.0 5.0 } 0.6 0.0 ee 10.0 5.0 0.6 0.2 0.0 10.0 5.0 1.2 | 0.6 0.6 10.0 5.0 0.9 | 0.8 0.3 5.0 2.5 0.8 | 0.0 et 5.0 2.5 1.3 0.4 0.7 5.0 2.5 1.6 1.2 1.0 5.0 2.5 2.0 1.6 1.4 10.0 5.0 0.8 0.5 0.2 10.0 5.0 } 0.9 0.5 0.3 10.0 5.0 0.8 0.5 0.2 10.0 5.0 0.8 0.0 = 10.0 5.0 0.8 0.2 0.2 10.0 5.0 1.0 0.6 0.4 10.0 5.0 1.3 0.8 0.7 5.0 2.5 0.6 0.0 = 5.0 2.5 1.0 0.4 0.4 5.0 2.5 1.4 1.2 0.8 5.0 2.5 2.0 1.6 1.4 10.0 5.0 0.5 0.0 == 10.0 5.0 1.4 0.2 0.8 10.0 5.0 0.9 0.4 0.3 10.0 5.0 1.1 0.6 0.5 10.0 5.0 0.4 0.0 ae 10.0 5.0 0.5 0.0 a 10.0 5.0 | 0.8 0.1 0.2 10.0 5.0 | 0.9 0.2 0.3 10.0 5.0 jt 0.3 0.5 10.0 5.0 0.8 0.4 0.2 10.0 5.0 1.0 0.5 0.4 10.0 5.0 1.5 0.6 0.9 * The phosphate-starch mixture was not arsenic-free. The average of 10 determinations made as above without arsenic addition was 0.6 part per million. This amount was subtracted from the total arsenic trioxid found to obtain the amount of added arsenic trioxid that was recovered. ® To each of the Jast 12 samples 0.5 gram of sodium chlorid was added. 518 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 Work was also carried out on the estimation of arsenic in gelatin, but the results do not justify a report at this time. The work is still in progress, as is that on a general hydrochloric acid distillation method which would be applicable to all foods. Two recommendations adopted by the association in 1915 were: “That the gravimetric and volumetric methods for tin, tested this year, be adopted by the association as provisional. “That further study be made of other methods for the determination of tin.” For reasons indicated above, it was not possible to carry out the pro- posed work on these two recommendations. Work is now in progress on other methods for the determination of tin, especially a modification of Parry’s method as used by Dr. W. B. D. Penniman. RECOMMENDATIONS. It is recomended— (1) That the revised Gutzeit procedure and its application to baking materials be made the subject of collaborative work for the year 1917 with a view to its provisional adoption in 1917. (2) That further study be made of the application of the method to gelatin and other food products. (3) That the provisional gravimetric and volumetric methods for the determination of tin be subjected to further study with a view to final adoption in 1917. (4) That further study be made of other methods for the determina- tion of tin. (5) That the methods for the determination of copper, zinc, nickel and aluminium in food products be made the subject of study by the association as soon as possible. W. D. Collins (Bureau of Chemistry, Washington, D. C.) presented a paper on ““C. R. Smith’s Method for the Determination of Arsenic’. The meeting adjourned at 5.07 p. m. for the day. 1J. Ind. Eng. Chem., 1918, 10: 360. THIRD DAY. WEDNESDAY—MORNING SESSION. REPORT OF COMMITTEE A ON RECOMMENDATIONS OF REFEREES AND REVISION OF METHODS. By W. W. Skinner (Bureau of Chemistry, Washington, D. C.), Chairman. [Phosphoric acid (basic slag, to cooperate with committee on vegetation tests on the availability of phosphoric acid in basic slag), nitrogen (special study of Kjeldahl method), potash, soils (nitrogenous compounds, lime requirements), inor- ganic plant constituents, insecticides and fungicides, and water.| PHOSPHORIC ACID. It is reeommended— (1) That the study of the preparation of neutral ammonium citrate solution, its use in determining reverted phosphoric acid and possible substitutes for it in this determination, be continued. Approved. (2) That in view of the conditions resulting from the European war, whereby the price of molybdic acid has been more than quadrupled and 100 per cent molybdic acid practically removed from the United States markets, the referee study the determination of phosphoric acid with a view to recommending an optional method not requiring the use of molybdic acid. Approved. (3) That the volumetric method, dissolving the slag in sulphuric and nitric acids!, be adopted as an official method for total phosphoric acid in basic slag. Approved. (4) That this association instruct its referee on phosphoric acid to give prominent attention to the question of methods of determining available phosphoric acid in slags, the chemical ingredients influencing the same, and the bibliography on the subject. Sufficient reports are already in the hands of your committee to be of service to the referee on phosphoric acid in his chemical investigations. It seems unnecessary to the committee to wait until all of the vegetation results are at hand before tentative methods of analysis are submitted to the association. Approved. 1J. Assoc. Official Agr. Chemis(s, 1917, 3: 90. 519 520 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 NITROGEN. It is recommended— (1) That the tentative ferrous-sulphate-zinc-soda method be adopted as official. Approved. (2) That, owing to the conflicting results on previous work, the use of glass wool in the neck of the distillation flask receive further study. Approved. (3) That further study be made of the effect of permanganate at the end of the digestion in the Kjeldahl modified method when used on a nitrate salt. Approved. (4) That the use of sodium sulphate in the Gunning method in place of potassium sulphate be tried out on a variety of organic substances of known origin, as well as of difficult oxidation. Approved. (5) That no further investigation of the Jones and Street methods for the determination of organic nitrogen activity be made. Approved. POTASH. It is reeommended— (1) That the work on the availability of potash be continued. Approved. (2) That the referee next year study further the barium hydroxid process of the perchlorate method on mixed fertilizers of known potash content. Approved. (3) That the official method for the preparation of solution! be revised to read as follows: Place 2.5 grams of the sample upon a 12.5 cm. filter paper and wash with successive portions of boiling water into a 250 cc. flask until the filtrate amounts to about 200 ce. Add to the hot solution a slight excess of ammonium hydroxid and sufficient ammonium oxalate to precipitate all of the lime present, cool, dilute to 250 cc., mix, and pass through a dry filter. After a lengthy discussion, the proposed revision was disapproved by the association, and the matter referred to the referee for next year. SOILS. It is recommended— (1) That further study be made of methods for total sulphur in soils, including a comparison of the following methods: Sodium peroxid fusion; heating soil with magnesium nitrate solution, as used for total phosphorus 1Assoc. Official Agr. Chemists, Methods, 1906, 12, 41 (a). 1920) SKINNER: RECOMMENDATIONS OF REFEREES 521 in soils; modification of Eschka’s method for sulphur in coal; ignition of soil with mixtures of magnesium oxid, sodium carbonate and am- monium nitrate. Approved. (2) That methods for the determination of the total constituents of soils be studied with a view to substituting them for the “strong acid digestion”! and be referred to the referee for next year. Approved. (3) That the recommendation of the committee in 1915 that a modifica- tion of the Marr method be made a provisional method? be reconsidered, and that the adoption of a method for the determination of inorganic carbon in soils be held in abeyance, pending further investigation. _ Approved. INORGANIC PLANT CONSTITUENTS. It is reeommended— (1) That the methods as outlined for calcium, magnesium, iron and aluminium’ be further studied on solutions approximating the com- position of the ash from cereals. Approved. (2) That the colorimetric method for the determination of manganese be further studied. Approved. INSECTICIDES AND FUNGICIDES. It is recommended— (1) That the method for the determination of arsenic trioxid in lead arsenate‘ be adopted as a tentative method. Approved. (2) That the method for the determination of arsenic pentoxid in lead arsenate® be adopted as a tentative method. : Approved. (3) That further study be made of methods for the determination of copper, lead and zinc in such preparations as Bordeaux-lead arsenate, Bordeaux-zinc arsenite, etc. Approved. (4) That further comparison be made of the zinc chlorid and iodin methods in the analysis of lime sulphur solutions. Approved. 1Asscc. Official Agr. Chemists, Methods, 1916, 22, 11. *J. Assoc. Official Agr. Chemists, 1917, 3: 60. 4Ibid., 1920, 3: 329. 4TIbid., 1920, 3: 332. + Ibid., 1920, 3: 363. 522 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 The foregoing recommendation, a substitute for the recommendation of the referee, resulted in considerable discussion, and a motion by Mr. Doolittle that the zinc chlorid method be made a tentative method and that the referee be directed to continue the study of the same, was duly seconded and carried. (5) That cooperative work be done on the Gyory method for titrating As" in hydrochloric acid solution with a solution of potassium bromate. Approved. (6) That Method I for total arsenic oxid! be dropped. Approved. (7) That the methods for the determination of moisture, free acetic acid and free ammonia proposed in 1910? and adopted as official (final action in 1912) be dropped. Approved. (8) That all other methods for insecticides and fungicides be adopted as tentative and official methods, as given in the Association of Official Agricultural Chemists, Methods, 1916, VII, 63-77, except as further modified in the 1916 Report of the Referee on Insecticides. Approved. Nore.—Mr. Skinner explained that the last above-mentioned recommendation was made for the purpose of covering certain points in the methods which have been slightly modified. WATER. It is recommended— (1) That the following method for the determination of lithium, potassium and sodium be adopted in 1917 as an official method: Dissolve the total alkali chlorids in a minimum amount of cold water in a tall 200 ce beaker (1.5 cc. will be more than sufficient for 0.5 gram of the salts). Add 1 drop of concentrated hydrochloric acid and gradually 20 cc. of absolute alcohol, dropping the alcohol into the center of the beaker (not on the sides) while rotating. Precipitate the sodium and potassium chlorid in a perfectly uniform granular condition. In a similar manner, while rotating the beaker, add 60 cc. of ether (sp. gr. 0.716-0.717) and allow the mixture to stand about 5 minutes, or until the precipitate is well agglomerated and the supernatant liquid almost clear. Occasionally rotate the beaker. Filter the mixture through a weighed Gooch crucible into an Erlenmeyer flask, using a bell-jar arrangement. Thoroughly wash the beaker with a mixture of 1 part of alcohol and 4-5 parts of ether. A rubber-tipped rod is necessary for this purpose. Also thoroughly wash the precipitate in the Gooch crucible and set the crucible aside. Thoroughly wash the funnel to remove any lithium therefrom into the flask containing the filtrate. Svaporate the filtrate to dryness on the steam bath (using a blast). Take up the residue with 10 cc. of absolute alcohol, warming if necessary, so that practically every- thing passes into solution. If a slight film remains on the bottom of the flask and sides, remove it by rubbing with a rubber-tipped glass rod. While rotating the flask, U.S. Bur. Chem. Bull. 107, rey.: 28. *Thid., 137: 38. ‘Thid., 162: 49. 4J. Assoc. Official Agr. Chemists, 1920, 3: 331. , ; 1920) SKINNER: RECOMMENDATIONS OF REFEREES 523 add 50 ce. of ether (sp. gr. at 35°C., 0.716-0.717). Add 1 drop of concentrated hydro- chloric acid, rotate the flask and allow to stand for 30 minutes. It is well to rotate the flask at frequent intervals. When the fine precipitate has agglomerated (only a very small amount is usually precipitated), filter it into a tall beaker through the same crucible as used in the first precipitation, a bell-jar arrangement being employed. Wash the residue with ether-alcohol mixture, using the same precautions as outlined in the first precipitation. After drying in an oven, gently ignite the crucible, cool, and weigh. Evaporate the ether-alcohol solution of lithium on the steam bath. Take up the residue in a little water and add a slight excess of sulphuric acid. Carefully transfer the solution to a weighed porcelain or platinum dish. Evaporate the solution as far as possible on the same steam: bath and gently ignite the residue over a flame. By placing the dish on a triangle over an asbestos gauze and using a low flame, the solu- tion can be evaporated without spattering. Cn Then carefully ignite the residue over a full flame. When charring has occurred, it is well to repeat the ignition with sulphuric acid. Calculate to lithium, using the factor 0.12625. Remove the chlorid of sodium and potassium from the Gooch crucible with 25-50 ce. of hot water, collecting the filtrate in a porcelain dish by meaus of the bell-jar arrange- ment. Add sufficient platinic chlorid solution (containing the equivalent of 1 gram of metallic platinum, i. e., 2.1 grams H.PtCl, in every 10 cc.) to convert sodium and potas- sium to their respective double chlorids and evaporate to dryness. Treat the residue with 80% alcohol, filter, and wash until the excess of platinic chlorid and sodium platinic chlorid has been removed. Dry the filter and precipitate, dissolve the residue in hot water, and transfer to a weighed platinum dish. Evaporate on the steam bath, dry for 30 minutes in the oven at 100°C. and weigh as potassium platinic chlorid; calculate to potassium chlorid, using the factor 0.30673, and to potassium, using the factor 0.16085. Find the weight of sodium chlorid by subtracting the weight of potassium cblorid from the total weight of the sodium chlorid and potassium chlorid obtained above. Calculate to sodium, using the factor 0.39343. Report as mg. sodium, mg. potassium, and mg. lithium per 50 cc. of solution. Approved. (2) At the suggestion of Mr. R. E. Doolittle of the Committee on Editing Methods of Analysis, certain methods have been recommended for adoption as official methods. These methods have been published in the Association of Official Agricultural Chemists, Methods, 1916, IV, but are not methods that have received recommendation for final adoption. In order to have the record straight, the referee recommends the adoption of the following methods as official: (a) Method for turbidity, (a) and (b)'. (b) Method for color, 3 and 4'. (ec) Method for odor, 5%. (d) The Schulze-Trommsdorf method for the determination of required oxygen, 22 and 237. (e) Method I and Method II for dissolved oxygen, 24, 25, 26 and 27%. 1 Assoc. Official Agr. Chemists, Methods, 1916, 35. 2Thid., 39. 3 Tbid., 40. 524 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (f) Method for the determination of specific gravity, 301. (g) Method for the determination of hydrogen sulphid, 37'. (h) Method for temporary hardness, 702. (i) Method for alkalinity, 71, 72, 73 and 742. (j) Method for total hardness, 75 and 76°. (k) Method for permanent or non-carbonate hardness, 77°. Adopted as official. (8) That the method for free carbon dioxid‘ remain a tentative method. Approved. (4) That the Gutzeit method for the determination of arsenic® be printed in the methods for the analysis of water as an optional official method. After some discussion a motion to postpone action on this last re- commendation for another year was carried. REPORT OF COMMITTEE B ON RECOMMENDATIONS OF REFEREES AND REVISION OF METHODS. By R. E. Srauurnes® (State Department of Agriculture, Atlanta, Ga.), Chairman. [Foods and feeding stuffs (sugar, crude fiber, stock feed adulteration, organic and inorganic phosphorus, water), dairy products (separation of nitrogenous sub- stances in milk and cheese), saccharine products (maple products, honey, sugar house products), drugs (medicinal plants, alkaloids, synthetic products, medicated soft drinks, balsams and gum resins, enzyms), testing of chemical reagents and micro-analytical methods. | FOODS AND FEEDING STUFFS. It is recommended— (1) That a further study of sulphur dioxid in bleached grains be made by collaborators. Approved. (2) That the method for determining the acidity of corn, as described by Black and Alsberg’, be considered by the referee next year with a view to its adoption as an official method, and that the method be studied to see if changes are necessary to make it applicable to grains other than corn. Approved. 1 Assoc. Official Agr. Chemists, Methods, 1916, 41. 2 Thid., 50. 8 Ibid., 51. ‘ Ibid., 42, 38. 6 Thid., 171. * Since deceased. Bur. Plant Ind. Bull. 199. ou 1920) STALLINGS: RECOMMENDATIONS OF REFEREES 52: SUGAR. It is recommended— (1) That the modifications proposed last year for determining sucrose by acid and invertase inversion be further studied. Approved. (2) That the work upon determining small amounts of reducing sugars in the presence of sucrose be continued. Approved. (3) That the methods of determining copper by reduction of the oxid in alcohol vapors be investigated. Approved. (4) That the optical methods for estimating raffinose in beet products be examined with special reference to hydrolysis by means of enzyms. ' Approved. (5) That details of mixing raw sugars be studied with a view to reducing moisture changes. Approved. (6) That the influence of temperature upon polarization by sugars other than sucrose be studied. Approved. (7) That recommendations 2, 3 and 5, made by W. D. Horne’, be referred to the Committee on Editing Methods of Analysis. Approved. (8) That the referees continue in collaboration with the Bureau of Standards the preparation of a table of reduction factors for the more common reducing sugars. Approved. CRUDE FIBER. It is recommended— (1) That the one filtration method? be investigated further. Approved. (2) That the matter of a uniform filtering medium be studied further. Approved. STOCK FEED ADULTERATION. It is reeommended— (1) That samples be sent out the coming year for cooperation in the determination of grit and weed seeds in scratch feeds. Approved. (2) That a key or outline for the qualitative detection of adulterants in feeding stuffs be prepared and submitted at the next meeting. Approved. 1 J. Assoc. Official Agr. Chemists, 1919, 3: 263. * Ibid., 256. 526 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 4 (3) That the following recommendations of 1914 be studied during the coming year: (1) That methods for the detection of peat, dried at high tempera- tures, in feeding stuffs be investigated. Approved. (2) That the maximum percentage of foreign materials permissible in mill by-products be investigated. Approved. ORGANIC AND INORGANIC PHOSPHORUS. It is reeommended— (1) That the magnesia mixture method! for the estimation of water- soluble inorganic phosphorus in flesh be adopted as an official method of this association, with one minor change of detail in the interest of economy of reagents, namely, that the amount of magnesia mixture used in extracts from 10 to 12 gram samples be reduced from 50 to 10 ce. The committee feels that sufficient collaboration has not been reported and that final action should be deferred until further collaboration. Final action postponed. (2) That further work be done with the magnesia mixture method! on brain; that other glandular tissues be studied. Approved. WATER. It is recommended— (1) That the referee for the ensuing year study further methods for determining water in foods and feeding stuffs, especially the use of the vacuum method with calcium oxid for cereal products in comparison with the official methods. Approved. DAIRY PRODUCTS. It is recommended— (1) That further study be made on the Harding-Parkin method for fat determination? in comparison with the present official and provisional methods. Approved. (2) That further study be given to enzym reactions of milk. Approved. 1 J. Assoc. Official Agr. Chemists, 1916, 1: 562; 1919, 3: 264. 2J. Ind. Eng. Chem., 1913, 5: 131. 1920] STALLINGS: RECOMMENDATIONS OF REFEREES 527 SEPARATION OF NITROGENOUS SUBSTANCES. It is reeommended— (1) That the referee for next year attempt to determine the relative amounts of some of the dissociation products in water-soluble and water-insoluble meat proteins. Approved. (2) That study be continued leading to the adoption of methods for the determination of the non-casein proteins and the products of protein decomposition in milk. Approved. SACCHARINE PRODUCTS. No report or recommendations. DRUGS. It is recommended— (1) That comparative work be resumed on the ricin method for the assay of pepsin! and that the methods outlined for the identification and essay of papain be studied cooperatively. Approved. (2) That the appointment of the referee on balsams be continued, and that a study be made of the methods of demonstrating the difference between the natural and the artificial product. Approved. (3) That the methods for the determination of strychnin in tablet triturates? be made provisional. Approved. (4) That the method for the determination of strychnin in liquids? where it occurs as the only alkaloid be made provisional. Approved. (5) That a further study be made of the method for determining atropin in tablets’. Approved. (6) That the work on alkaloids be extended to a study of methods for the determination of strychnin and quinin in admixture. Approved. (7) That the work on mixtures containing synthetic products be continued. Approved. ' Assoc. Official Agr. Chemists, Methods, 1916, 363. 2 J. Assoc. Official Agr. Chemists, 1919, 3: 189; 1920, 3: 379. * Thid., 1920, 3: 379. 528 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 TESTING OF CHEMICAL REAGENTS. It is recommended— (1) That the work on the determination of alcohol in pharmaceutical preparations be continued. Approved. (2) That the method for the determination of the strength of acetic anhydrid! be studied cooperatively. Approved. (3) That the work on tests for purity of immiscible organic solvents be taken up. Approved. MICRO-ANALYTICAL METHODS. New subject. REPORT OF COMMITTEE C ON RECOMMENDATIONS OF REFEREES AND REVISION OF METHODS. By H. E. Barnarp? (State Board of Health, Indianapolis, Ind.), Chairman. [Food preservatives, coloring matters in foods, metals in foods, fruit and fruit products, canned vegetables, cereal foods, wines, soft drinks (bottlers’ products), distilled liquors, beers, vinegars, flavoring extracts, meat and meat products (separa- tion of nitrogenous compounds in meat products, meat extract dairy products, edible fats and oils, spices and other condi- ments, cacao products, coffee, tea, baking powder.] FOOD PRESERVATIVES. It is recommended— (1) That Method I, as given in the report of the referee on preserva- tives’, for the determination of saccharin in the absence of other ether- soluble sulphur compounds be adopted as a tentative method to replace the method given in Bureau of Chemistry Bulletin 107 (Revised), page 183, and in the Association of Official Agricultural Chemists, Methods, 1916, 145. Approved. (2) That the Jorissen test for salicylic acid, as given in the report of the referee on preservatives’, be adopted as a tentative method. Approved. (8) That further work be done on Method II, as given in the report of the referee on preservatives®, for the determination of saccharin in 1 Acetic anhydrid was treated in the cold with anilin, as described by Menschutkin and Vasilieff (J. Russ. Phys. Chem. Soc., 1889, 21: 190). The acetanilid formed was weighed. 2 Present address, American Institute of Baking, Minneapolis, Minn. 3 J. Assoc. Official Agr. Chemists, 1920, 3: 504. 4 Ibid., 512. 6 Tdid., 505, 1920] BARNARD: RECOMMENDATIONS OF REFEREES 529 the presence of mustard oil; that other methods not dependent upon the sulphur component of saccharin be investigated; and that further work be done upon the determination of saccharin in baked flour preparations. Approved. (4) That the following methods be made official, the paragraph num- bers and titles being given as they appear in the Association of Official Agricultural Chemists, Methods, 1916, 141-54: SALICYLIC ACID. 1 PREPARATION OF SAMPLE.—OFFICIAL. 2 Ferric Chlorid Test.—Qualitative—Official. 4 Colorimetric Method.—Quantitative-—Official. BENZOIC ACID. 6,7 PREPARATION OF SAMPLE.—OFFICIAL. 9 Ferric Chlorid Test.—Qualitalive —Official. 10 Modified Mohler Test.—Qualitative-—Offcial. 11 Quantilalive Method.—Official. SACCHARIN. 12 Qualitative Test.—Official. : BORIC ACID AND BORATES. 14 Qualitative Test.—Official. 15 Quantilalive Method.—Offcial. FORMALDEHYDE. 16 PREPARATION OF SAMPLE.—OFFICIAL. 17 Phenylhydrazin Hydrochlorid Method.—Official. 18 Hehner Method.—Offictal. 19 Leach Method.—Official. 20 Phenylhydrazin Hydrochlorid and Sodium Nitro-prussid Test.—Official. 21 Phenylhydrazin Hydrochlorid and Potassium Ferricyanid Test.—Official. 22 Phenylhydrazin Hydrochlorid and Ferric Chlorid Test.—Offcial. 23 Phloroglucol Method.—Official. FLUORIDS. 24 Method I.—Modified Method of Blarez.—Official. 25 Method II.—Official. SULPHUROUS ACID. 30 Method I.—Distillation Method.—Official. 31 Method II.—Direct Titration Method.—Official. 32 DETERMINATION OF FREE SULPHUROUS ACID.—OFFICIAL. FORMIC ACID. 38, 39, 40 Quantitative Method.—Official. Approved. 530 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (5) That the other methods in this chapter remain tentative. Approved. COLORING MATTERS IN FOODS. No recommendations. METALS IN FOODS. It is recommended— (1) That the revised Gutzeit procedure and its application to baking materials be made the subject of collaborative work for the year 1917. Approved. (2) That further study be made of the application of the method to gelatin and other food products. Approved. (3) That the tentative gravimetric and volumetric methods for tin be subjected to further study. Approved. (4) That further study be made of other methods for the determina- tion of tin. Approved. (5) That the methods for the determination of copper, zinc, nickel and aluminium in food products be made the subject of study. Approved. FRUIT AND FRUIT PRODUCTS. It is recommended— (1) That the methods for the determination of malic acid! be adopted as tentative methods. Approved. (2) That the method for the determination of citric acid® be adopted as a tentative method. Approved. CANNED VEGETABLES. It is recommended— That the Howard method as amended for the examination of tomato pulp and its products be retained as a tentative method. The method is as follows: 49 APPARATUS. (a) Compound microscope-—Equipped with apochromatic objectives and compen- sating oculars, giving magnifications of approximately 90, 180, and 500 diameters. These magnifications can be obtained by the use of 16 and 8 mm. Zeiss apochromatic 1J. Assoc. Official Agr. Chemists, 1919, 3: 403-5. 2 Thid., 405-6. 1920| BARNARD: RECOMMENDATIONS OF REFEREES 531 objectives with X6 and X18 Zeiss compensating oculars, or their equivalents, such as the Spencer 16 and 8 mm. apochromatic objectives with Spencer X10 and X20 com- pensating oculars, the drawtube of the microscope being adjusted as directed below. (b) Thoma-Zeiss blood counting cell. (C) Howard mold counting cell.—Constructed like a blood counting cell but with the inner disk (which need not be ruled) about 19 mm. in diameter. 50 MOLDS.—TENTATIVE. Clean the special Howard cell so that Newton’s rings are produced between the slide and the cover glass. Remove the cover and place, by means of a knife blade or scalpel, a small drop of the sample upon the central disk; spread the drop evenly over the disk and cover with the cover glass so as to give an even spread to the material. It is of the utmost importance that the drop be mixed thoroughly and spread evenly; otherwise the insoluble matter, and consequently the molds, are most abundant at the center of the drop. Squeezing out the more liquid portions around the margin must be avoided. In a satisfactory mount Newton’s rings should be apparent when finally mounted and none of the liquid should be drawn across the moat and under the cover glass. Place the slide under the microscope and examine with a magnification of about 90 diameters and with such adjustment that each field of view represents approxi- mately 1.5 sq. mm. of area on the mount. This area is of vital importance and may be obtained by adjusting the drawtube to the proper length as determined by actual measurement of the field, a 16 mm. Zeiss apochromatic objective with a Zeiss X6 compensating ocular or a Spencer 16 mm. apochromatic objective with a Spencer X10 compensating ocular, or their equivalents, being used to obtain the proper magnification. Observe each field as to the presence or absence of mold filaments and note the result as positive or negative. Examine at least 50 fields, prepared from two or more mounts. No field should be considered positive unless the aggregate length of the filaments present exceeds approximately one-sixth the diameter of the field. Calcu- late the proportion of positive fields from the results of the examination of all the observed fields and report as percentage of fields containing mold filaments. 51 YEASTS AND SPORES.—TENTATIVE. Fill a graduated cylinder with water to the 20 cc. mark, and then add the sample till the level of the mixture reaches the 30 cc. mark. Close the graduate, or pour the contents into an Erlenmeyer flask, and shake the mixture vigorously 15-20 seconds. To facilitate thorough mixing, the mixture should not fill more than three-fourths of the container in which the shaking is performed. For tomato sauce or pastes, or products running very high in the number of organisms, or of heavy consistency, 80 cc. of water should be used with 10 cc. or 10 grams of the sample. In the case of exceptionally thick or dry pastes, it may be necessary to make an even greater dilution. Pour the mixture into a beaker. Thoroughly clean the Thoma-Zeiss counting cell so as to give good Newton's rings. Stir thoroughly the contents of the beaker with a scalpel or knife blade, and then, after allowing to stand 3-5 seconds, remove a small drop and place upon the central disk of the Thoma-Zeiss counting cell and cover immediately with the cover glass, observing the same precautions in mounting the sample as given under 50. Allow the slide to stand not less than 10 minutes before beginning to make the count. Make the count with a magnification of about 180, to obtain which the following combinations, or their equivalents, should be employed: 8 mm. Zeiss apochromatic objective with X6 Zeiss compensating ocular, or an 8 mm. 532 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 Spencer apochromatic objective with X10 Spencer compensating ocular with draw- tube not extended. Count the number of yeasts and spores on one-half of the ruled squares on the disk (this amounts to counting the number in eight of the blocks, each of which contains twenty-five of the small ruled squares). The total number thus obtained equals the number of organisms in «oc cmm. if a dilution of 1 part of the sample with 2 parts of water is used. If a dilution of 1 part of the sample with 8 parts of water is used, the number must be multiplied by 3. In making the counts, the analyst should avoid count- ing an organism twice when it rests on a boundary line between two adjacent squares. 52 BACTERIA. TENTATIVE. Estimate the bacteria from the mounted sample used in 51, but allow the sample to stand not less than 15 minutes after mounting before counting. Employ a magnifi- cation of about 500, which may be obtained by the use of an 8 mm. Zeiss apochro- matic objective with an X18 Zeiss compensating ocular with drawtube not extended, or an 8 mm. Spencer apochromatic objective with an X20 Spencer compensating ocular having a tube length of 190, or their equivalents. Count and record the number of bacteria in a small area consisting of five of the small-sized squares. Move the slide to another portion of the field and count the number on another similar area. Count five such areas, preferably one from near each corner of the ruled portion of the slide and one from near the center. Determine the average number of bacteria per area and multiply by 2,400,000, which gives the number of bacteria per ce. If a dilution of 1 part of the sample with 8 parts of water, instead of 1 part of the sample with 2 parts of water, is used in making up the sample, then the total count obtained as above must be multiplied by 7,200,000. Omit the micrococci type of bacteria in making the count. Approved. CEREAL FOODS. It is recommended— That the following methods be studied during the coming year: (1) Moisture-—Comparison of the official method with the vacuum method, using calcium oxid. (2) Gluten.—Comparison of methods of washing gluten by using (a) distilled water; (b) water containing sodium chlorid; (c) ordinary hydrant water. (3) Soluble carbohydrates—Comparison of methods using different strengths of hydrochloric acid. (4) Cold water ectract. (5) Chlorin. Approved. WINES. It is recommended (1) That the method suggested by the associate referee for 1915! for determining the acidity in wines, and discussed by the referee for 1916, be adopted as tentative. Approved. 1 J. Assoc. Official Agr. Chemisls, 1917, 2: 186. 2 Tbid., 1920, 3: 409. 1920] BARNARD: RECOMMENDATIONS OF REFEREES 533 (2) That the following methods be studied during the coming year: (a) Determination of tartaric acid present as esters by saponify- ing before determining the total tartaric acid. (b) Determination of the acidity of red wines by the clarification method'. (c) Determination of glycerol according to the Rothenfusser method?. Approved. SOFT DRINKS. New subject. No report. DISTILLED LICUORS. No recommendations. BEERS. No recommendations. VINEGARS. No recommendations. FLAVORING EXTRACTS. It is reecommended— (1) That the methods of analysis for imitation vanilla preparations containing large quantities of vanillin and coumarin be given further study. Approved. (2) That the advisability of making a preliminary test for coumarin in vanilla extracts be studied. Approved. (3) That the value of the test for the detection of vanilla resins be studied. Approved. (4) That the applicability of Hortvet and West’s method for deter- mining alcohol in lemon and orange extracts be studied. Approved. (5) That Mitchell’s polarization method be studied for the purpose of determining the most desirable factors to be used, especially with _ reference to the natural variation in the oils and the influence of dilution. Approved. (6) That Albright’s details of the Kleber method for citral in lemon and orange oil® be studied. Approved. 1J. Assoc. Official Agr. Chemists, 1920, 3: 410. 2Z. Nahr. Genussm., 1912, 23: 332-7. 2 J. Assoc. Official Agr. Chemists, 1920, 3: 417. 534 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (7) That available and new methods for determining benzoic acid in almond extract be studied. Approved. (8) That Howard’s method! and the present tentative method of determining the oil in cassia, coumarin and clove extracts be further studied. Approved. MEAT AND MEAT PRODUCTS. No report. DAIRY PRODUCTS. It is reeommended— (1) That the modifications of the Roese-Gottlieb method applied to plain ice cream, dried milk and malted milk be further studied. Approved. (2) That the Schmidt-Bondzynski modified method for the deter- mination of fat in cheese be adopted as a tentative method, and further studied. Approved. (3) That Patrick’s method for the determination of sucrose in sweetened condensed milk, as outlined below, be adopted as a tenta- tive method: PATEIN AND DUFAU’S REAGENT’. To 220 grams of yellow mercuric oxid and 300-400 cc. of water in an evaporating dish, add cautiously sufficient concentrated nitric acid (about 140 cc.) just to dissolve the mercuric oxid, then add a solution of sodium hydroxid in sufficient quantity to give a slight permanent precipitate, dilute to 1 liter and filter. (As this reagent tends to become more acid with age, through the deposition of basic salts of mercury, it should receive the addition of a little alkali from time to time.) DETERMINATION. To 50 grams of the 20 per cent solution, or 25 grams of a 40 per cent solution, of the sample in a 100 ce. graduated flask, add 25 cc. of water, then 5 cc. of the Patein and Dufau reagent, and shake well. Without delay run in, with constant shaking, sufficient N/2 sodium hydroxid to make the mixture practically neutral, but not alkaline, to litmus paper (12 to 13 cc. of N/2 sodium hydroxid is usually required; the amount should be determined beforehand on 5 cc. of the Patein and Dufau reagent). Make up to the mark with water, shake well, filter, and polarize ina 200 mm. tube at 20°C. Invert the sucrose in 50 ce. of this solution by adding 5 cc. of concentrated hydro- chloric acid and letting stand overnight at room temperature (above 20°C.). Obtain the invert reading at 20°C. without neutralizing the acid and multiply by 1.1 to correct for dilution. 1920] BARNARD: RECOMMENDATIONS OF REFEREES 535 A correction of the direct reading, and a further correction of the invert reading, for the volume occupied by the fat and proteins, using the factor 1.075 for fat and 0.80 for protein, should be made. After making these corrections, calculate the sucrose by the Clerget formula: 100 (P—I). L SS in which 142.35 — — 2 S=per cent of sucrose; P=direct reading in degrees Ventzke; I =invert reading; T =temperature at which polarization was made. Approved. EDIBLE FATS AND OILS. It is recommended— (1) That the method reported by the referee!, for the detection of beef fat in lard by crystallization from acetone and the preparation and determination of the melting point of the fatty acids, be further studied with a view to its adoption as a tentative method in 1917. Approved. (2) That the potassium-salt-acetone method? for the separation of solid and liquid fatty acids be given further study. Approved. SPICES AND OTHER CONDIMENTS. It is reecommended— (1) That the modification of the distillation method for water in whole spices, as reported by the associate referee, be given further study with particular reference to the size and dimensions of the apparatu and length of time of heating. (2) That the subject of sampling and grinding and the preparation for analysis of each spice be studied. COCOA AND COCOA PRODUCTS. It is recommended— (1) That the name of this subject be changed from “Cocoa and Cocoa Products” to “Cacao Products’. Approved. (2) That the corrected formula for the polariscopic determination of sucrose and lactose be adopted as tentative. Approved. 1 J. Assoc. Official Agr. Chemists, 1920, 3: 433. 2 [bid., 435. 536 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (3) That the tentative method for the determination of fat! be changed to provide for a 4-hour extraction. Approved. (4) That the proposed modification of the Baier and Neumann method be further studied. Approved. (5) That the determination of the critical temperature of dissolution for the examination of cacao butter be further studied with a view to its adoption as a tentative method. Approved. (6) That the associate referee’s test for tallow and hydrogenated oils be further studied. Approved. TEA AND COFFEE. It is recommended— That the Stahlschmidt method for caffein be further studied with a view to its adoption as an official method. Approved. BAKING POWDER. It is reeommended— (1) That the Exner method for the gravimetric determination of lead (now a tentative method) be dropped, and that no further study of it be made. Approved. (2) That a further study be made of the Wichmann method and modifications for its improvement. Approved. (3) That a study be made of Bryan’s modification of the Corper method for the electrolytic determination of lead in baking powder. Approved. (4) That a study be made of Chittick’s method for the determination of lead. Approved. 1 Assoc. Official Agr. Chemists, Methods, 1916, 328. ¢ 1920| REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 537 REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS}. Your Committee on Editing Methods of Analysis submitted at the meeting of the association last year a tentative draft of the revised methods. This draft was submitted in order that the members of the association might have an opportunity to study the changes made by the addition of new methods, deletion of obsolete and incorrect methods, rearrangements, changes in phraseology, etc. As a result, many volun- tary criticisms and suggestions were received. The revised methods were also submitted to some of the members who were particularly familiar with their history, development and adaptability, in order to insure their correctness. As a result of the information thus secured, a considerable number of changes were introduced and the revised methods forwarded to the secretary of the association for publication. Errors, however, have slipped into the published methods due to oversight on the part of the committee and through editorial changes introduced into the text without participation on the part of the committee, which have necessitated a careful review of the methods as printed in the Association of Official Agricultural Chemists, Methods, 1916. The report as offered this year includes many more alterations than the committee would desire, but in fairness to the association it seems necessary that we should offer them for your consideration in a complete form. The committee desires to make the following general recommendations: (1) That the following listed general reference tables be placed by themselves in a separate chapter at the end of the methods and desig- nated as Chapter XXX, and that appropriate changes be made in the marginal numbers and in the cross references to them in the text of the methods: 1 Munson and Walker’s Table?. For calculating dextrose, invert sugar alone, invert sugar in the presence of sucrose, etc. 2 Krdber’s Table*. For determining pentoses and pentosans. 3 Table for the densities of solutions of cane sugar at 20°C.+ 4 Table of temperature corrections for changing percentages of sugar by specific gravity to true values at 20°C.°® 5 Geerlig’s Table®. For dry substances in sugar-house products by the Abbé refrac- tometer, at 28°C. 6 Table of corrections for temperature to be used in conjunction with Table No. 5’. 1Presented by R. E. Doolittle. 2 Assoc. Official Agr. Chemists, Methods, 1916, 88-96. 3 Tbid., 112-7. 4 Ibid., 125-6. 5 U.S. Bur. Standards Circ. 19: 1916. 25. ® Assoc. Official Agr. Chemists, Methods, 1916, 127 7 [bid., 128. 538 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 7 Alcohol Table’. For calculating the percentages of alcohol in mixtures of ethyl alcohol and water from their specific gravities. 8 Alcohol Table?. For calculating the percentages of alcohol in mixtures of ethyl alcohol and water from their Zeiss immersion refractometer readings at 17.5-25°C. 9 Table of international atomic weights, 1916. Approved. (2) That the word “‘chlorin” wherever it appears in the expressions ‘washed free from chlorin” and “washed practically free from chlorin” be changed to “‘chlorids”’. Approved. (3) That the gravimetric factors be restored throughout the text of the book of methods. Approved. At this point a motion was made, seconded and adopted that the By-laws be sus- pended. (4) That, where the strength of solutions is given in terms of per cent, specific instructions be substituted giving the manner in which the solu- tions are to be prepared. Approved. (5) That all alkaloids be spelled with the final “‘e”’. After considerable discussion a substitute motion was introduced recommending that alkaloids be spelled without the final “e”. The substitute motion was duly seconded and adopted. The committee desires to make the following specific recommendations: I. FERTILIZERS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricultural Chemists, Methods, 1916, 1-15, be changed as follows, and as changed be adopted as the official and tentative methods of the association for the analysis of fertilizers: CHANGES. (1) 5, PREPARATION OF SOLUTION. Combine the first and the last paragraphs and change to read as follows: ‘“Treat 2 grams of the sample by one of the methods given below. In the case of (d) 2.5 grams may be used. Cool, dilute to 200 cc., or to 250 cc. if a 2.5 gram sample was used. Mix, and pour on a dry filter.” Approved. 1 Assoc. Official Agr. Chemists, Methods 1916, 194-207. 2 [bid., 208-35. 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 539 (2) 6, DETERMINATION. Line 5.—Eliminate “‘60-80” and substitute therefor “50°, making the sentence read: “To the hot solution add 50 cc. of the molybdate solution for every decigram of phosphoric acid (P20;) that is present.” Disapproved. Mr. F. P. Veitch (Bureau of Chemistry, Washington, D. C.) made a motion that “70” be substituted for “60-80”, making the sentence read: “To the hot solution add 70 cc. of the molybdate solution for every decigram of phosphoric acid (P20;) that is present.” Approved. Line 12.—Eliminate the words ‘‘add magnesia mixture’’. Approved. , Line 13.—After “vigorously” insert ‘‘15 cc. of magnesia mixture for each decigram of phosphoric acid (P:0;) present,” making the corrected sentence read: “Nearly neutralize with hydrochloric acid, cool, and from a burette add slowly (about 1 drop per second), stirring vigorously, 15 cc. of magnesia mixture for each decigram of phos- phoric acid (P:0;) present.” Approved. Line 16.—Eliminate the remainder of the paragraph beginning with the words “dry the filter’ and substitute therefor the words “ignite to whiteness or to a grayish white, weigh and calculate to phosphoric acid (P,Os).’’, making the latter part of the paragraph read: “Let stand till the supernatant liquid is clear (2 hours is usually enough), filter, wash with the dilute ammonium hydroxid until the washings are practically free from chlorids, ignite to whiteness or to a grayish white, weigh and calculate to phosphoric acid (P20;).” Approved. (3) 9, DETERMINATION. Line 14.—Change the word “a’’ before the word “beaker” to “the’’, making the sentence read: “Transfer the precipitate and filter to the beaker or precipitating vessel,”’. Approved. (4) 11, Volumetric Method.—Oficial. Line 2.—Eliminate the words ‘and ammonium hydroxid until a slight permanent precipitate is formed” and substitute therefor “nearly neutralize with ammo- nium hydroxid’’, making the sentence read: “‘add 10 cc. of concentrated nitric acid, nearly neutralize with ammonium hydroxid, dilute to 60 cc., and proceed as directed under 9.” Approved. (5) 12, nEAGENts. Line 5.—Eliminate the words “litmus or azolitmin paper” and substitute therefor “a saturated alcoholic solution of corallin’’, making the parenthetical expression read, “(testing with a saturated alcoholic solution of corallin)”. Approved. 540 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (6) 13, perERMINaTION. Line 2.—Eliminate the word ‘‘Erlenmeyer’’, making the sentence read: ‘Heat 100 ec. of strictly neutral ammonium citrate solution (sp. gr. 1.09) to 65°C. in a 250 ce. flask placed in a warm water bath,”. Approved. (7) 21, DETERMINATION. Line 7.—Eliminate the words “Do not add either potassium permanganate or po- tassium sulphid. Cool, dilute, neutralize, distil, and titrate with the standard alkali. In neutralizing”, and substitute therefor “Complete the determination, as directed under 18, except that neither potassium permanganate nor potassium sulphid is added. In making alkaline’, making the sentences read: “Digest for a time after the mixture is colorless or nearly so, or until oxidation is complete. Complete the determination, as directed under 18, except that neither potassium permanganate nor potassium sulphid is added. In making alkaline before distilling, it is convenient to add a few drops of phenolphthalein indicator, etc.” Approved. (8) 27, REAGENTS AND APPARATUS. Line 1.—Insert the word “apparatus” after the word “the’’, making the sentence read: ‘The apparatus, reagents and standard solutions are described under 16, 17, 19 and 24.” Approved. (9) 32, Magnesium Oxid Method —Offcial. At the end of the paragraph add the words “using cochineal or methyl red solution as indicator’, making the sentence read: “and titrate with standard alkali solution, using cochineal or methyl red solution as indicator.” (10) 34, Zine-Iron Method.—Official. At the end of the paragraph add the words “using cochineal or methyl red solution as indicator’, making the sentence read: ‘Continue the distillation until 100 cc. have been distilled and titrate with standard alkali solution, using cochineal or methyl red solution as indicator.” (11) 35, Ferrous Sulphate-Zinc-Soda Method.—Tentative. At the end of the paragraph add the words “using cochineal or methyl red solution as indicator”, making the sentence read: “‘and titrate with standard alkali solution, using cochineal or methyl red solution as indicator.” Recommendations 9, 10 and 11 approved. (12) 38, PREPARATION OF SAMPLE. Line above this heading.—Eliminate the parenthetical expression “(Not applicable to fertilizers containing cottonseed meal or castor pomace.)”’. Approved. (13) 39, peTERMINATION. Line 2.—Before the word “Kjeldahl” insert the word “round-bottomed”. Approved. V 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 541 (14) 40 (c), 80% alcohol. Eliminate the paragraph as printed and subsitute therefor the following: “(C) 80% alcohol—Sp. gr. 0.8593 at ~» Approved. Attention was called to the fact that if each change made by the committee were to be passed upon separately it would take several days to dispose of the report of the committee. It was then decided that Mr. Doolittle should bring only questions involving actual changes in the methods to the attention of the association. (15) 41 (a), Mized fertilizers. Line 2.—After the word “with” insert “successive small portions of”, making the sentence read: “‘and wash with successive small portions of boiling water’. (16) 41 (c), Organic compounds. _Line 3.—Change “250” to “500”. Line 4.—Eliminate the words “‘and proceed as in (€)’’ and substitute therefor the words “cool, dilute to 500 cc., mix, pass through a dry filter and proceed as in 42 (a)”’, making the last sentence of this paragraph read : “Add a little strong hydrochloric acid, warm slightly in order to loosen the mass from the dish, transfer to a 500 cc. graduated flask, add ammonium hydroxid and ammonium oxalate, cool, dilute to 500 cc., mix, pass through a dry filter and proceed as in 42 (a).”’ (17) Add a new paragraph after 41 (C) for the preparation of the solution of wood ashes, cotton hull ashes and similar materials, as follows: “(d) Ashes from wood, cotton hulls, etc—Boil 10 grams of the sample with 300 ce. of water for 30 minutes, add a slight excess of ammonium hydroxid to the hot solution and then sufficient ammonium oxalate to precipitate all of the lime present. Cool, dilute to 500 cc., mix, pass through a dry filter and proceed as in 42 (a).” (18) 42 (d), Water-soluble potash in wood ashes and cotton hull ashes. Line 1.—Change this title to agree with 41 (d), viz.: “(d) Water-soluble potash in ashes from wood, cotton hulls, etc.” Line 2.—Change “(a)” after “41” to “(d)”, making the reference read “Al (d)”. (19) Between 42 (d) and Method II, Official, insert the following paragraph: “For the conversion of potassium platinic chlorid to potassium chlorid use the factor 0.3067; to potassium sulphate 0.3585; to potassium oxid 0.1938.” (20) 48, pREPARATION OF SOLUTION. Line 1.—Change “‘(e)” to “(&)”, making the reference read “5 (8)”. (21) 50, Volumetric Method. Line 1.—Change “(e)” to “(&)”, making the reference read “5 (8)”. Methods on fertilizers, as changed, adopted as a whole. II. SOILS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricultural Chemists, Methods, 1916, 17-28, be changed as follows, and as changed 542 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 be adopted as the official and tentative methods of the association for the analysis of soils: CHANGES. (1) 2}. PREPARATION OF SAMPLE.—OFFICIAL. Line 1.—Insert “(a)” at the beginning of the first paragraph. Line 2.—Eliminate the words ‘‘to avoid” and substitute therefor the words, ‘“‘and avoiding”, making the sentence read: “After air-drying and weighing the sample, pulverize in a porcelain mortar, using a rubber-tipped pestle and avoiding the reduc- tion of rock fragments, and pass through a sieve with circular openings =s inch (1 mm.) in diameter”. Line 6.—Insert ‘(b)” at the beginning of the second paragraph. Line 6.—Eliminate the word ‘quantitative’ and after the word “determination” insert the words “‘of the total quantity’, making the paragraph read: “(b) For the determination of the total quantity of any of the constituents, etc.” (2) Change the reference number ‘“‘2” appearing in line 1 of paragraphs 3, 6, 9,10, 11 and 28 to “2 (a)”. (3) Change the reference number “2” in 23, line 1, and 24, line 2, to “2 (b)”, and insert the reference “2 (b)” after the word ‘‘soil’” in 25, line 2. (4) 5 (b), FIG. 2, PARR’S APPARATUS FOR THE DETERMINATION OF CARBON DIOXID. Insert the letter “B” to designate the gas burette. (5) 8, APPARATUS. Line 7.—After the word “‘tube”’ insert “‘of the same diameter’’, making the sentence read: “‘which in turn is furnished with a side tube of the same diameter extending through the condenser jacket (D)”’. (6) 9, DETERMINATION. Line 11.—Eliminate the words “‘and most of the lower large bulb”. (7) 12, INSOLUBLE RESIDUE.—OFFICIAL. Last line of paragraph.—Eliminate ‘(a)’ and substitute therefor the words, “be- ginning with ‘Wash the residue from the filter’ ”’, making the sentence read: “and com- plete the determination as directed under III, 4, beginning with ‘Wash the residue from the filter’ ” (8) 13, IRON, ALUMINIUM AND PHOSPHORIC ACID, COLLECTIVELY.—OFFICIAL. Line 14.—Fliminate the word “filtrate”? and substitute therefor the words ““com- bined filtrates’, making the sentence read: ‘‘Designate the combined filtrates as B’’. (9) 14, MANGANESE.—OFFICIAL. Line 3.—Eliminate the second “‘of”’ and the preceding comma and substitute therefor the word “and”, making the sentence read: “repeat the addition of bromin water and ammonium hydroxid and boil again.” (10) 16, MAGNESIUM.—OFFICIAL. Line 10.—After the expression “[I, 4 (d)]” eliminate the rest of the paragraph and substitute therefor the words: “dry, burn first at a moderate heat, then ignite intensely and weigh as magnesium pyrophosphate (Mg.P,0;), and calculate to mag- nesium oxid (MgQ).” 1920| REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 543 (11) 18 (a), Standard sodium or potassium hydrozid solution. Line 1.—After “solution.—” insert the clause, “Prepare a solution of such strength that 100 cc. exactly neutralizes 16.19 cc. of normal acid;’’ and eliminate the words “strength such that”, making the sentence read: “Prepare a solution of such strength that 100 cc. exactly neutralizes 16.19 cc. of normal acid; 1 cc. of this solution is equiva- lent to 0.0005 gram of phosphorus pentoxid (P.0;).” (12) 23, Magnesium Nitrate Method.—Official. Line above this heading.—Change the title “Total Phosphorus” to “Total Phosphoric Acid.” (13) 25, TOTAL POTASSIUM.—OFFICIAL. Line 25.—Elminate the words “by means of suction” and after “‘wash’’ insert ‘‘en- tirely free of soluble platinic salts”, making the sentence read: ‘‘Filter through a small filter, wash entirely free of soluble platinic salts with 80% alcohol, then with ammonium chlorid solution (I, 40 (a)], and finally with 80% alcohol.” ‘ Line 26.—Eliminate the sentence reading: “‘Dry the precipitate on the filter and wash the precipitate with hot water into a weighed platinum dish, using suction.”’ and substitute therefor the following: ‘Dry the precipitate on the filter, then dissolve and wash the precipitate through the filter with hot water into a weighed platinum dish.” (14) 27, PHOSPHORUS SOLUBLE IN N/5 NITRIC ACID.—TENTATIVE. Change the title “Phosphorus Soluble in N/5 Nitric Acid.—Tentative” to “Phos- phoric Acid Soluble in N/5 Nitric Acid.—Tentative.” (15) 28, cALcluM CARBONATE REQUIRED.—TENTATIVE. Line 6.—Change “‘0.001” to “0.01”, and after the word “used” insert “assuming that the total amount of acid present is 2.5 times the amount in the solution titrated’, making the parenthetical phrase read: ‘‘(0.01% on basis of the weight of soil used, assuming that the total amount of acid present is 2.5 times the amount in the solution titrated)”. (16) 29, STATEMENT OF RESULTS.—OFFICIAL. Transpose the line reading “Organic carbon’ to a position immediately below “Volatile matter”. Change the line reading “Inorganic carbon” to ‘Carbon dioxid (CO:) equivalent of inorganic carbon.”” After the words “Volatile matter” insert “other than carbon dioxid’’. Lines 18 and 20, change the word ‘“‘phosphorus” to “phosphoric acid”. This will change the column after the line reading ‘Sulphur trioxid (SO,)”’ to read: “Carbon dioxid (CO) equivalent of inorganic carbon.................-. Volatile matter other than carbon dioxid...................--.00-2-05. UGRIT GET Iie oe Ba Sa ORES ce IOC EES Bee ORD COROT OSEARIS tite: es HEN MTOPEN ee orator tof. <. « APA T SEIS yale, SAAR os eee sels ois 5 stags atotevslors PR eet tep HOS PMOL ACI sys ses «soo eye a ous sks ote teeusia\s foo etaigiela alate aioe ores opens MES EMSUGLARSHUINIM EN: Fite cif Via: slacisil. siee ge ptardee Ohisedeie aes bbs Phosphoric acid soluble in N/5 nitric acid.............-..-.-.---4---- i AleNunrCATvOnAte LEQUILEG. <7). = . Eliminate the last two lines, which read: “The pycnometer determination should not 17.5°C. 20°C. ;, be made at any other temperature than [7~=> or qe (4) 9, Tasie 11.—Densities of solutions of cane sugar at 20°C. Eliminate the parenthetical statement in two places, “(This table is the basis for standardizing hydrometers indicating the per cent of sugar at 20°C.)”. Transfer this table to Chapter X XX, Reference Tables, and designate as “3”, and add as “4” the correction table given on page 25, Bureau of Standards Circular 19 (1916), entitled “Temperature Corrections to Readings of Saccharometers (standard at 20°C.)”’. 554. ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (5) 10, REFRACTOMETER METHOD.—TENTATIVE. Line 2.—Eliminate ‘‘11” and substitute therefor “XXX, 5”. Eliminate “12” and substitute therefor “XXX, 6”. Line 3. (6) 11, Tasie 12.—Geeruic’s TABLE. Transfer to Chapter X XX, Reference Tables, and designate as ‘5’. (7) 12, Taste 13.—Correctlions for temperature. Transfer to Chapter X XX, Reference Tables. and designate as “6”. (S) 31, ALCOHOL IN SIRUPS USED IN CONFECTIONERY (“BRANDY DROPS”).—TENTATIVE Line 9.—Eliminate “XVI, 5”, and substitute therefor ‘“X XX, 7”. (9) 35, moIsTURE. Insert the word “‘Official’’ at the end of the line, making the heading read, *“Moisture.—Official.”’ (10) After 58 insert a paragraph, 59, as follows: 59 COMMERCIAL GLUCOSE.—TENTATIVE. Proceed as directed under 25. Nore: This will necessitate renumbering the remaining headings of the chapter. (11) BIBLIOGRAPHY. Reference 9.—Change “‘p. 60” to “‘p. 59.” Reference 15.—Change “p. 17” to “p. 21.” Methods on saccharine products, as changed, adopted as a whole. X. FOOD PRESERVATIVES. That the revised methods, as reported by the Committee on Editing Methods of Analysis and published in the Association of Official Agricul- tural Chemists, Methods, 1916, 141-54, be changed as follows, and as changed be adopted as the tentative methods of the association for the determination of preservatives in foods: CHANGES. (1) 1 (a), Non-alcoholic liquids. Line 2.—After the word “substances” insert the words “which cause troublesome emulsions during extractions”, making the sentence read: “If gums or mucilaginous substances which cause troublesome emulsions during extraction are present, pipette 100 ce. into a 250 cc. volumetric flask, etc.” ’ (2) 4, ExTRACTION. Line 9.—Eliminate the word ‘‘an” before the word “emulsion” and substitute there- “a small amount of”. Line 10.—After the word “layer” insert the phrase “where it is frequently broken during the next extraction’, making the sentence read: “If a small amount of emulsion still persists, allow it to remain with the aqueous layer, where it is frequently broken during the next extraction.” for the words ee eee r 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 555 (3) 13, Quantitative Method. Line 15.—Eliminate the words “over an alcohol or other sulphur-free flame” and sub- stitute therefor the sentences: ‘“The fusion must be conducted so that gases contain- ing sulphur do not come into contact with the melt. This can be accomplished by using an alcohol flame (Barthel burner) or, if illuminating gas is used, by fitting the crucible into a piece of heavy asbestos board, so that the upper third of the crucible projects above the board, the lower portion of the crucible being in contact with the flame.” (4) 16, PREPARATION OF SAMPLE. The second paragraph reading: “‘In the case of meats * * * * and filter from any insoluble matter” is applicable only to the qualitative tests for formaldehyde. Therefore transfer this paragraph as the first paragraph under 17. (5) 20, Phenylhydrazin Hydrochlorid and Sodium Nitro-prussid Test. Line above this heading—Change “‘Rimini Method” to “Rimini’s Methods.” Insert “I” before ““Phenylhydrazin Hydrochlorid and Sodium Nitro-prussid Test.” (6) 21, Phenylhydrazin Hydrochlorid and Potassium Ferricyanid Test. Insert “II” before ““Phenylhydrazin Hydrochlorid and Potassium Ferricyanid Test.” (7) 22. Phenylhydrazin Hydrochlorid and Ferric Chlorid Test. Insert “III” before ““Phenylhydrazin Hydrochlorid and Ferric Chlorid Test.” Methods on food preservatives, as changed, adopted as a whole. XI. COLORING MATTERS IN FOODS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 155-69, be changed as follows, and as changed be adopted as the tentative methods of the association for the determination of coloring matters in foods: CHANGES. (1) 6, ORANGE I AND ERYTHROSINE. Line 1.—After the word “‘extract’’ insert “from which some of the colors"have been removed,” making the sentence read: ‘‘Measure, if necessary, the amyl alcohol extract from which some of the colors have been removed, under 5, then’’. (2) 7, INDIGO CARMINE, AMARANTH AND TARTRAZINE. Lines 16, 19, 22.—Change the word “hyposulphite”’ to ‘““hydrosulphite”’. (3) 15, SPECIAL TESTS FOR COAL TAR DYES PERMITTED UNDER THE FEDERAL FOOD AND DRUGS ACT Lines 5, 13, 15, 16, and paragraph 5, line 2.—Change the word “‘hyposulphite” to “hydrosulphite”’. (4) 20 (c), Sodium hyposulphite solution. Lines 1 and 2.—Change the word “‘hyposulphite” to “hydrosulphite’’, making the paragraph read: “(C) Sodium hydrosu!phite solution.—A freshly prepared 5% solution of ‘Blankite’, sodium hydrosulphite (Na2S20,).”” 556 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (5) 21, Sodium hyposulphile. Line 7.—Change the word “‘hyposulphite” to “hydrosulphite” (two places), making the line read: ‘Sodium hydrosulphite-—Add the sodium hydrosulphite solution drop by drop.” (6) 23, CHLOROPHYLL. Last line of paragraph.—Eliminate the words “returning to green in a few minutes” and substitute therefor “quickly returning to green’’, making the sentence read: ““The color becomes brown, quickly returning to green.” (7) 26, COCHINEAL. Line 13.—Eliminate the words “‘so sensitive to small amounts” and substitute there- for the words “‘so characteristic’, making the sentence read: “This, however, is not so characteristic as the first test and many fruit colors give tests hardly to be distinguished.” Methods on coloring matters in foods, as changed, adopted as a whole. XII. METALS IN FOODS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 171-6, be changed as follows, and as changed be adopted as the tentative methods of the association for the determination of metals in foods: CHANGES. (1) 1 (b), Sulphuric acid (1 to 2). Line 1—After the word “acid”? insert “‘arsenic-free’’, making the line read, “(b) Sulphuric acid, arsenic-free, (1 to 2).” (2) 4, pereRMENATION. At the end of the paragraph add the following: “‘Conduct a blank test on the reagents alone and correct the result for any arsenic so found. The blank should not exceed 0.001 mg.” (3) 5, Gravimetric Method.—Tentative. At the end of the last paragraph add the words “using the factor 0.7881", making the last sentence read: ‘Weigh as stannic oxid and calculate to metallic tin, using the factor 0.7881.” (4) 9, ZINC.—TENTATIVE. At the end of the paragraph add the words “using the factor 0.8034", making the last sentence read: ‘Calculate the weight of metallic zinc, using the factor 0.8034.” Methods on metals in foods, as changed, adopted as a whole. ou ou “I 1920| REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS XII. FRUITS AND FRUIT PRODUCTS. That the revised methods, as reported by the Committee on Editing Methods of Analysis in the Association of Official Agricultural Chemists, Methods, 1916, 177-84, be changed as follows, and as changed be adopted as the official and tentative methods of the association for the analysis of fruits and fruit products: CHANGES. (1) 8, SULPHATE AND CHLORID.—TENTATIVE. Change the title to read: “Sulphates and Chlorids.—Tentative.” At the end of the first paragraph add “using the factor 0.7465", making the last sentence of the first paragraph read: “From the weight of barium sulphate calculate the sulphate present as per cent of potassium sulphate, using the factor 0.7465.” (2) 9, TOTAL ACIDITY.—TENTATIVE. Line 7.—After the word “‘in” insert “‘per cent or’, making the phrase read, “‘expressing the results in per cent or grams per 100 cc.” (3) 19, Qualitative Test.—Tentative. Eliminate the first three sentences and substitute therefor the following: “Dilute a portion of the sample with water, heat nearly to boiling, add several cc. of dilute sulphuric acid, and then add potassium permanganate solution until all color is destroyed. Cool and test with iodin solution.” (4) 25, Method I—Tentative. Insert reference mark after “Method I’, and place in bibliography the reference “U.S. Bur. Chem. Circ. 76.” Line 2.—Eliminate ‘‘and a dichromate cell’. Lines 22 and 23.—Eliminate ‘“‘and with a dichromate cell”. (5) 28, pREPARATION OF SOLUTION. Line 7.—Eliminate the words “‘in 100 cc.” (6) 31, pETERMINATION. Line 12.—Change **250” to ‘*400”. Line 16.—Change the sentence beginning “After removing from the bath” to read: “After removing from the bath, add rapidly from a burette 25 cc. of the 5% potassium permanganate solution, drop by drop with frequent interruptions, and with constant, vigorous shaking, avoiding a temperature during oxidation exceeding 55°C.” Methods on fruits and fruit products, as changed, adopted as @ whole. XIV. CANNED VEGETABLES. That the revised methods reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 185-6, be changed as follows, and as changed be adopted as the official and tentative methods of the asso- ciation for the analysis of canned vegetables: 558 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 CHANGE. 7, TOTAL ACIDS.—TENTATIVE. Line 1.—Eliminate the sentence: ‘Express the result as citric acid; 1 cc. of N/10 alkali is equivalent to 0.0070 gram of crystallized citric acid” and substitute therefor the sentence: “Express the results as number of cc. of N/10 alkali required to neutralize 100 grams of sample.” Methods on canned vegetables, as changed, adopted as a whole. XV. CEREAL PRODUCTS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 187-91, be changed as follows, and as changed be adopted as the official and tentative methods of the asso- ciation for the analysis of cereal products: CHANGES. (1) 2, ASH.—OFFICIAL. Line 1.—After the word “‘using’’ insert “3—’, making the phrase read, “using 3-5 grams of the flour.” (2) 8, Method I. (By nitrogen determination)—Tentative. Line 5.—Eliminate the words ‘‘allow to settle’, making the sentence read: “Shake thoroughly once more and filter through a dry, folded filter, returning the first runnings to the filter until a clear filtrate is obtained.” (3) 11, PROTEIN SOLUBLE IN 5 PER CENT POTASSIUM SULPHATE SOLUTION.—TENTATIVE. Line 3.—Eliminate the words “‘let stand overnight”. Line 3.—Change the expression “3 hours” to “1 hour’’, making the corrected sen- tence read: ‘‘Shake at 30 minute intervals for 3 hours or, better still, agitate at moderate speed in a shaker for 1 hour, let settle 30 minutes, filter, etc.” (4) 14, COLD WATER-SOLUBLE EXTRACT.—TENTATIVE. Line 2.—After “10°C.” insert the words ‘or lower’’, making the phrase read, “200 ce. of water at 10°C. or lower”. Last line of paragraph.—Transpose the words “in an oven at 100°C. for periods of 30 minutes” and place after the word “dry”, making the sentence read, ‘‘and dry, in an oven at 100°C. for periods of 30 minutes, to constant weight.” (5) 18, Quantitative Method. (Added Chlorin in Chlorin-Bleached Flours) —Tentative. Line 6.—After the word “sodium” before the word “‘hydroxid” insert the words “‘or of potassium”, making the sentence read: ‘25 grams of sodium or of potassium hydroxid and 15 grams of sodium nitrate per liter.” At the end of this paragraph add the following as a new paragraph: “Special pre- cautions should be taken that the air of the laboratory during the entire operation is not contaminated by chlorin or hydrochloric acid fumes and that all reagents employed are as free as possible from chlorin. In all cases a blank determination should be con- ducted at the same time and correction introduced if necessary.” Methods on cereal products, as changed, adopted as a whole. 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS — 559 XVI. WINES. That the revised methods, as reported by the Committee on Editing Methods of Analysis and published in the Association of Official Agricul- tural Chemists, Methods, 1916, 193-242, inclusive, be changed as follows, and as changed be adopted as the official and tentative methods of the association for the analysis of wines: CHANGES. (1) XVI. WINES. Immediately under the title insert the following parenthetical expression, ‘‘(Unless otherwise noted, express results as grams per 100 cc.)”’. (2) 3, SPECIFIC GRAVITY._TENTATIVE. Line 1.—At the end of the sentence add: “Standardize the pycnometer as follows: Carefully clean the pycnometer by filling with a saturated sclution of chromic acid in concentrated sulphuric acid and allowing to stand for several hours. Empty the pyc- nometer and rinse thoroughly with water. Then fill it with recently boiled water, previously cooled to 16—18°C., place in a bath of water cooled to the same temperature and allow to warm slowly to 20°C. When the temperature has reached exactly 20°C., strike off the level of the water at the proper point on the pycnometer with a piece of filter paper, adjust the cap in place, remove from the bath, wipe dry with a cloth and, after allowing to stand for 15-20 minutes, weigh. Empty the pycnometer, rinse several times with alcohol and then with ether, allow it to become perfectly dry and weigh. Ascertain the weight of contained water at 20°C. by subtracting the weight of the empty pycnometer from its weight when full, and calculate the weight of contained water at 4°C. by multiplying the results by 1.0018 (determined from the respective aa 1.00000 densities of water at the two temperatures ooo): “To determine the specific gravity of the wine at ae cool the latter to 16-18°C., fill the pycnometer. immerse in a water bath cooled to 16-18°C., allow to warm slowly to 20°C., strike off at the mark, adjust the cap, wipe dry and weigh exactly as de- scribed above for standardization with water. Subtract the weight of the empty pycnometer from its weight when filled with the wine, and divide the difference by the weight of contained water at 4°C. determined above, the quotient being the 5 . 3 20265, specific gravity of the wine at —~. Mr. H. B. McDonnell (Agricultural Experiment Station, College Park, Md.) made the following recommendation, which was accepted by the Committee on Editing Methods of Analysis and adopted by the asso- ciation: 3, SPECIFIC GRAVITY.—TENTATIVE. That the cap of the pycnometer have a capillary opening at the top, or near the top, to provide for the expansion of the liquid. (8) 4 (a), By volume. Line 10.—Eliminate “5” and substitute therefor “XXX, 7”. 560 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (4) 4 (b), Grams per 100 ce. Line 2.—Eliminate ‘‘5” and substitute therefor “XXX, 7”. (5) 4 (d), By immersion refractometer. Line 3.—Eliminate ‘6” and substitute therefor “XXX, 8”. (6) 5, TasBLe 16.—ALcoHOoL TABLE. Transfer this table to Chapter XXX, Reference Tables, and designate as “7, Alcohol Table.” (7) 6, TasLe 17.—Axconor TABLE. Transfer this table to Chapter X XX, Reference Tables, and designate as “8, Alco- hol Table.” Mr. S. H. Ross (Bureau of Chemistry, Washington, D. C.) made the following recommendation, which was seconded by Mr. E. G. Grab (Bureau of Chemistry, Washington, D. C.), and adopted by the associa- tion: 8, Method II. (By Oxidation with Dichromate)—Tentative. Line 5.—After the word “‘stirring” insert “‘except that before the addition of the silver carbonate the residue is transferred with hot water to a 100 cc. graduated flask’. The last portion of the paragraph will then read as follows: “Proceed from this point as directed under XTX, 6, beginning with the clause ‘evaporate almost to dryness, with frequent stirring’, except that before the addition of the silver carbonate the residue is transferred with hot water to a 100 cc. graduated flask. Observe the pre- cautions given concerning the temperature at which all evaporations are to be made.” (8) 11, From the Specific Gravity of the Dealcoholized Wine.—Tentative. Line 7.—Eliminate “TX, 9” and substitute therefor “XXX, 3”. (9) 14 (b), Sweet wines. Line 4.—Change ‘“‘245” to “240”. (10) 23, suLPHURIC ACID.—TENTATIVE. Last line of paragraph.—After “*(SOs)”’ insert “using the factor 0.3430". Methods on wines, as changed, adopted as a whole. XVII. DISTILLED LIQUORS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and published in the Association of Official Agricul- tural Chemists, Methods, 1916, 243-8, inclusive, be changed as follows, and as changed be adopted as the official and tentative methods of the association for the analysis of distilled liquors: CHANGES. (1) il. SPECIFIC GRAVITY.—TENTATIVE. Line 1.—After the word “pycnometer” insert “‘as directed under XVI, 3”. 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 561 Line 1.—After the word “‘or’ er “by means of’’, making the paragraph read: “Determine the specific gravity at => Coby means of a pycnometer as directed under XVI, 3, or by means of a small, peeeaeeal graduated hydrometer.” (2) 2, ALCOHOL BY WEIGHT.—OFFICIAL. Line 4.—Eliminate “XVI, 5” and substitute therefor “XXX, 7”. Line 8.—Eliminate “XVI, 6” and substitute therefor “XXX, 8”. (3) 3, Method I.—Official. Line 2.—Eliminate ‘““X VI, 5” and substitute therefor “XXX, 7”. (4) 4, Method IT.—Tentative. Line 3—Eliminate “XVI, 5” and substitute therefor “XXX, 7”. Line 7.—Eliminate “X VI, 6” and substitute therefor “XXX, &”. (5) 11 (a), Standard furfural solution. Line 1.—Omit ‘‘(@)”, so that the heading will read: ‘‘Standard furfural solution.” At the end of the paragraph add the sentence: “The strong furfural solution will retain its strength but the dilute solutions will not.” (6) 11 (b), Furfural-free alcohol. Eliminate the entire line. (7) 13 (a), Purified carbon tetrachlorid. Add as a second paragraph: ‘The refuse carbon tetrachlorid after titration is purified for further work by collecting in a large bottle, adding concentrated sodium bydroxid solution, shaking, washing with tap water until the washings are neutral to phenol- phthalein, and distilling.” Methods on distilled liquors, as changed, adopted as a whole. XVIII. BEERS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and published in the Association of Official Agricul- tural Chemists, Methods, 1916, 249-51, inclusive, be changed as follows, and as changed be adopted as the official and tentative methods of the association for the analysis of beers: CHANGES. (1) XVIII. BEERS. Immediately under the heading “X VIII. BEERS.” insert in small print, ‘(Unless otherwise noted, express results as grams per 100 cc.)”’. (2) 3, SPECIFIC GRAVITY.—TENTATIVE. Line 1.—After the word ‘ Rie insert “‘as eae under XVI, 3”, making the sentence read: as directed under XVI, 3.” (3) 7, Method IIT.—Tentative. Line 7.—Eliminate “IX, 9” and substitute therefor 2), ©, ©, Gian 562 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (4) 18, PROTEIN.—OFFICIAL. Line 3.—Before the word “multiply” insert the word ‘“‘and’’. Line 3.—Eliminate the words “‘and calculate the percentage of protein’’ and substi- tute therefor the words “to obtain the equivalent of protein”. This makes the corrected phrase read, “‘and multiply the result by 6.25 to obtain the equivalent of protein.” Methods on beers, as changed, adopted as a whole. XIX. VINEGARS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and published in the Association of Official Agricul- tural Chemists, Methods, 1916, 253-8, inclusive, be changed as follows, and as changed be adopted as the official and tentative methods of the association for the analysis of vinegars: CHANGES. (1) 3, SPECIFIC GRAVITY.—TENTATIVE. Line 1.—After the word “‘pycnometer”’ insert “as directed under X VI, 3”, making : : 3 20°C. - the sentence read: ‘Determine the specific gravity at a by means of a pycnome- ter, as directed under XVI, 3.” (2) 4, ALCOHOL.—TENTATIVE. Line 5.—Eliminate “XVI, 5” and substitute therefor “XXX, 7”. (3) it LEAD PRECIPITATE.—TENTATIVE. Line 1.—After the expression ‘20%’ insert “neutral’’, making the line read, ““To 10 cc. of the sample in a test tube, add 2 cc. of 20% neutral lead acetate solution”. Methods on vinegars, as changed, adopted as a whole. XX. FLAVORING EXTRACTS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 259-69, be changed as follows, and as changed be adopted as the official and tentative methods of the asso- ciation for the analysis of flavoring extracts: CHANGES. (1) 1, SPECIFIC GRAVITY.—TENTATIVE. Line 1.—At the end of the sentence add the words “as directed under XVI, 3”. (2) 4, prePARATION OF SOLUTION. Line 4.—After the expression “8%” insert the word “neutral’’. 70 Line 7.—FEliminate the word “normal”. 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 563 (3) 6, NORMAL LEAD NUMBER.—TENTATIVE. Change the title to read: “Lead Number.—Tentative.” Lines 6 and 9.—Eliminate the word “‘normal’’. (4) 7, TOTAL SOLIDS.—TENTATIVE. 8, ASH.—OFFICIAL. Change the word “‘grams’”’ to “‘cc.” (5) 17, SPECIFIC GRAVITY.—TENTATIVE. Line 1.—At the end of the sentence add the words “‘as directed under X VI, 3”. (6) 18, aLcoHOL.—TENTATIVE. Line 10.—Change the reference “XVI, 5” to “XXX, 7”. (7) 22 (a), Aldehyde-free alcohol. Line 2.—Insert after the word “‘for’’ the words “‘at least’’. (8) 23, DETERMINATION. Line 8.—After the expression “2 cc.”’ insert ‘‘(or a suitable amount)’’. (9) 32, SPECIFIC GRAVITY.—TENTATIVE. Line 1.—At the end of the sentence insert the words “as directed under XVI, 3.” Methods on flavoring extracts, as changed, adopted as a whole. XXI. MEAT AND MEAT PRODUCTS. That the methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricultural Chem- ists, Methods, 1916, 271-86, be changed as follows, and as changed be adopted as the official and tentative methods of the association for the analysis of meat and meat products: CHANGES. (1) 25, apparatus. Line 8.—After the word ‘“‘tubing’’ add the words “to prevent fracture, but arrange the latter so that it will not interfere with the free exit of gas.” (2) 26, pETERMINATION. Line 16.—Insert the phrase, ‘“Note the volume of nitric oxid contained in the tube, the temperature, and barometric pressure, and’’. Line 16.—Change ‘“‘Calculate”’ to “calculate”, making the sentence read: ‘‘Note the volume of nitric oxid contained in the tube, the temperature, and barometric pressure, and calculate the volume of nitric oxid at 0°C. and 760 mm. pressure.” Dr. P. F. Trowbridge (Agricultural Experiment Station, Agricultural College, N. Dak.) and Dr. F. C. Cook (Bureau of Chemistry, Washing- ton, D. C.) made the following recommendations, which were adopted by the association: 564 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS (1) 1, PREPARATION OF SAMPLE.—TENTATIVE. 2, MOISTURE.—TENTATIVE. 3, ASH.—OFFICIAL. 4, CRUDE FAT OR ETHER EXTRACT.—OFFICIAL. 5, TOTAL PHOSPHORUS.—TENTATIVE. Retain these paragraphs with the old headings and numbers. (2) 8, TOTAL NITROGEN.—OFFICIAL. Change “8” to “6”. (3) 14, apparatus. Change “14” to “7”. (4) 15, perennation. Change “15” to “8”. (5) 24, REAGENT. Change “24” to “9”. (6) 25, apparatus. Change “25” to “10”. (7) 26, pETERMINATION. Change “26” to “11”. (8) 27, REAGENTS. Change “27” to “12”. (9) 28, DETERMINATION. Change “28” to “13”. (10) 17, Qualitative Test—Tentative. Change “17” to “14”. (11) 18, Mayrhofer Method, Price Modification —Tentative. Change “18” to “15”. (12) 19, Qualitative Test—Tentative. Change “19” to “16”. (13) 20, prePARATION OF SOLUTION. Change “20” to “17”. (14) 21, pererMiNnation. Change “21” to “18”. (15) 22, rEAGENTs. Change “22” to “19”. (16) 23, pereRMINATION. Change “23” to “20”. (17) 29, PRESERVATIVES.—TENTATIVE. Change “29” to “21”. [Vol. III, No. 4 1920| REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 565 (18) 30, METALS.—TENTATIVE. Change “30” to “22”. (19) 31, coLORING MATTERS.—TENTATIVE. Change “31” to “23”. (20) 9, SOLUBLE AND INSOLUBLE NITROGEN.—TENTATIVE. Change “9” to “24”. Eliminate the heading ‘‘Soluble and Insoluble Nitrogen.—Tentative”’ and substitute therefor the general heading ““Water Extract.—Tentative.” and the subheading ‘‘Prep- aration of Solution.” Include in this paragraph all of the material given in 9 (old number) through the word “thoroughly”, fifth line from the end of the paragraph. The amended paragraph will then read as follows: 24 WATER EXTRACT.—TENTATIVE. PREPARATION OF SOLUTION. Exhaust 7-25 grams of the sample depending upon the water content in the follow- ing manner: Weigh into a 150 cc. beaker, add 5-10 ce. of cold (15°C.) ammonia-free water and stir to a homogeneous paste. Then add 50 cc. of cold water, stir every 3 minutes for 15 minutes, let stand for 2-3 minutes and decant the liquid upon a quanti- tative filter, collecting the filtrate in a 500 cc. graduated flask. Drain the beaker, pressing out the liquid from the meat residue by the aid of a glass rod. Add to the residue in the beaker 50 cc. of cold water, stir for 5 minutes, allow to stand 2-3 minutes, and decant as before. If a considerable portion of the meat is carried over onto the filter, transfer it back to the beaker by means of a glass rod. Repeat the extractions, using the following additional amounts of cold water: 50, 50, 25, 25, 25, and 25 ce. After the last extraction transfer the entire insoluble portion to the filter and wash with three 10 cc. portions of water, allowing the material to drain thoroughly after each addition of water. Dilute to the mark and mix thoroughly. (21) 10, conNECTIVE TISSUE NITROGEN.—TENTATIVE. Eliminate this paragraph and substitute therefor the following paragraph, number- ing it “25”: 25 SOLUBLE AND INSOLUBLE NITROGEN. Determine the total nitrogen in a 50 cc. aliquot of the solution obtained under 24, proceeding as directed under I, 18, 21 or 23. Subtract the percentage of soluble nitrogen from the percentage of total nitrogen, 6, to obtain the percentage of insoluble nitrogen. To obtain the percentage of insoluble protein, multiply the percentage of insoluble nitrogen by 6.25. (22) 11, coAGULABLE PROTEINS.—TENTATIVE. Change ‘‘11” to “26”, and change the heading “Coagulable Proteins.—Tentative.” to “Coagulable Nitrogen.—Tentative.” (23) 12, Modified Tannin-Salt Method.— Tentative. Line above this heading —Change to “‘Proteose, Peptone and Gelatin Nitrogen.”, and change “12” to “27”. (24) 13, MEAT BASES.—TENTATIVE. Change “13” to “28”. (25) 16, cREATIN.—OFFICIAL. Change “16” to “29”. 566 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (26) 44, REAGENTS. Change “44” to “30”. (27) 45, apparatus. Change “45” to “31”. (28) 46, pETERMINATION. Change “46” to “32”. (29) Add three new paragraphs, 33, 34 and 35, to read as follows: 33 Soérenson Formol Titration Method.—Tentative. To 20 cc. of the filtrate from 26, or 20 cc. of a solution containing an extract of the meat (in some cases a larger volume may be necessary), add 10 cc. of a freshly prepared phenolphthalein-formol mixture (50 cc. of commercial formol containing 1 cc. of a 0.5% solution of phenolphthalein in 50% alcohol, exactly neutralized with N/5 barium or sodium hydroxid). Titrate the mixture with N/5 barium hydroxid solution until a distinct red color appears, add a slight known excess of N/5 barium hydroxid and titrate back to neutrality with N/5 hydrochloric acid. Conduct a blank titration with the same reagents, using 20 cc. of water in place of the solution to be tested. From the amount of N/5 barium hydroxid required to neutralize the mixture, corrected for the amount used in the blank titration, calculate the amount of amino nitrogen present (including ammonia if this has not been removed). One cc. of N/5 barium hydroxid is equivalent to 2.8 mg. of amino‘nitrogen. ; 34 TOTAL SOLUBLE PHOSPHORUS.—TENTATIVE. Evaporate to dryness 50 cc. of the water extract prepared under 24, moisten the residue with 10 cc. of concentrated sulphuric acid, add a few drops of nitric acid and heat on a hot plate until all of the organic matter is destroyed. Add 100 cc. of water, boil for a few minutes and proceed as directed under I, 6. 35 SEPARATION OF SOLUBLE INORGANIC AND ORGANIC PHOSPHORUS.—TENTATIVE. To 500 cc. of the extract, prepared as directed under 24, add 50 cc. of magnesia mixture [I, 4 (C)] and proceed as directed under 37. (30) 6, PREPARATION OF SOLUTIONS. Line above this heading.—Eliminate this line and substitute therefor ‘‘Soluble Phos- phorus in Blood, Brain and Glandular Organs.—Tentative.” Change “6” to “36”. (31) 7, DETERMINATION. Change “7” to “37”. (32) 32, PREPARATION OF SAMPLE.—TENTATIVE. Change “32” to “38”. (33) MOISTURE.—TENTATIVE. Change “33” to “39”. (34) 34, ASH.—OFFICIAL. Change “34” to “40”. (35) 35, TOTAL PHOSPHORUS.—TENTATIVE. Change “35” to “41”. 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 567 (36) 36, CHLORIN.—TENTATIVE. Change “36” to “42”. (37) 37, FAT.—TENTATIVE. Change “37” to “43”. (38) 38, TOTAL NITROGEN.—OFFICIAL. Change “38” to “44”. (39) 41, AMMONIA.—TENTATIVE. Change “41” to “45”. Change this paragraph to read as follows: “Introduce 1 gram of pasty extracts or 2-3 grams of fluid extracts into tube (B) of the Folin apparatus and proceed as directed under 8.” (40) 39. INSOLUBLE PROTEIN.—TENTATIVE. Change this heading to read, “Insoluble Nitrogen.—Tentative.”, and change “39” to “46”. (41) 40, COAGULABLE PROTEIN.—TENTATIVE. Change this heading to. read, “‘Coagulable Nitrogen.—Tentative.’’, and change “40” to Oy ty FD (42) 42, PROTEOSES AND GELATIN.—TENTATIVE. Change “42” to “48”. (43) 43, GELATIN.—TENTATIVE. Change “43” to “49”. (44) Add a new paragraph, 50, to read as follows: 50 AMINO NITROGEN.—TENTATIVE. Proceed as directed under 32 or 33, using an aliquot of the filtrate from 47. (45) 47, ACID ALCOHOL-SOLUBLE NITROGEN.—TENTATIVE. Change “47” to “51”. (46) 48, CREATIN.—OFFICIAL. Change “48” to “52”. (47) 49, CREATININ.—OFFICIAL. Change “49” to “53”. (48) 50, Cook Method.—Tentalive. Change “50” to “55”. (49) 51, suGAR.—TENTATIVE. Change “51” to “56”. (50) 52, PRESERVATIVES.—TENTATIVE. Change “52” to “57”. (51) 53, METALS.—TENTATIVE. Change “53” to “58”. Methods on meat and meat products, as changed, adopted as a whole. 568 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. IJ, No. 4 XXII. DAIRY PRODUCTS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 287-98, be changed as follows, and as changed be adopted as the official and tentative methods of the associa- tion for the analysis of dairy products: CHANGES. (1) 8, REAGENTS. Line above this heading.—Change “Tentative” to “Official”. (2) 12, Roese-Gottlieb Method.—Official. Line 13.—After the word “‘parts’’ insert the words “‘free from suspended water’. (3) 16 (b), Ash. Line 8 —Eliminate the last sentence of paragraph and substitute therefor the fol- lowing: “The acetic serum ash multiplied by the factor 1.021 equals the sour serum ash (dilution of the acetic serum being 2%).” (4) 17 (a), Zeiss immersion refractometer reading. Line 1.—Insert bibliography reference “‘5’’ after “Zeiss immersion refractometer reading.” (5) 17 (b), Ash. Line 1.—Change ‘“‘Ash®” to “Ash°*”’. (6) BIBLIOGRAPHY. After line 5 add the following as a new reference: “®Z. offent. Chem., 1903, 9: 173.” Line 6.—Change bibliography reference “‘5” to “6”. Methods on dairy products, as changed, adopted as a whole. XXIII. FATS AND OILS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 299-315, be changed as follows, and as changed be adopted as the official and tentative methods of the asso- ciation for the analysis of edible fats and oils: CHANGES. (1)€Change the type in the title for 2, line above 3, titles for 5, 6 and 7, to conform with 11 (italics). (2) 2, at 20°C-__Tentative. Line 1.—After the word “‘pycnometer” add the words “‘as directed under XVI, 3.” 1920| REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 569 (3) 4, DETERMINATION. Line 1.—Eliminate the expression “dried at the temperature of boiling water’ and insert the word “dry” before the word “‘flask’’, making the phrase read: ‘“‘Fill the dry flask with the dry, hot, freshly filtered fat’’. (4) 5, General Directions.—Tentative. Last paragraph, line 1—Eliminate the word “‘directly”’ and after the word “gravity” add ‘and in the same direction.” (5) 15 (b), N/10 sodium thiosulphate. Line 1.—After the heading ““N/10 sodium thiosulphate.—”’ insert the sentence: “Dis- solve 24.82 grams of recrystallized sodium thiosulphate (Na:SO;.5H,O) in water and dilute to 1 liter.” The paragraph will then read: “(b) N/10 sodium thiosulphate— Dissolve 24.82 grams of recrystallized sodium thiosulphate (Na2SO;.5H2O) in water and dilute to 1 liter. Standardize this solution, etc.” (6) 15 (e), N/10 potassium dichromate. Line 1.—After the heading ““N/10 potassium dichromate.—” insert the sentence: “Dissolve 4.903 grams of potassium dichromate in water and dilute to 1 liter.” The paragraph will then read: ‘(€) N/10 potassium dichromate.—Dissolve 4.903 grams of potassium dichromate in water and dilute to 1 liter. The dichromate solution should be checked against pure iron.” ” (7) 1, REAGENTS. Change paragraph number “1” to “19”. (8) 24, saponrFICATION. At the end of the last paragraph add the sentence: “Remove the last traces of alcohol by waving the flask briskly, mouth down, or better, by a current of air free from carbon dioxid.” (9) 28, Polenske Method.—Tentative. Second paragraph, line 5.—Eliminate the words “‘as obtained above” and substitute therefor the words “‘obtained as above upon a 5 gram sample”. Methods on fats and oils, as changed, adopted as a whole. XXIV. SPICES AND OTHER CONDIMENTS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 317-26, be changed as follows. and as changed be adopted as the official and tentative methods of the asso- ciation for the analysis of spices and other condiments: CHANGES. (1) 24, sotips.—TENTATIVE. Line 3.—Eliminate the words “‘at 100°C.” 570 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IJ], No. 4 (2) 35, TOTAL SOLIDS.—TENTATIVE. Line 1.—Eliminate the word ‘‘platinum’’. (3) 49, apparatus. Line above this heading. —Eliminate the last three words and substitute therefor “Sauce and Paste’’. Eliminate (a), (b) and (C) and substitute therefor the following: (a) Compound microscope.—Equipped with apochromatic objectives and compen- sating oculars, giving magnifications of approximately 90, 180, and 500 diameters. These magnifications can be obtained by the use of 16 and 8 mm. Zeiss apochromatic objectives with X6 and X18 Zeiss compensating oculars, or their equivalents, such as the Spencer 16 and 8 mm. apochromatic objectives with Spencer X10 and X20 com- pensating oculars, the draw-tube of the microscope being adjusted as directed below. “(b) Thoma-Zeiss blood counting cell. “(C) Howard mold counting cell—Constructed like a blood counting cell but with the inner disk (which need not be ruled) about 19 mm. in diameter.” (4) 50, MoLDs.—TENTATIVE. Line 1.—Eliminate the expression ‘““Thoma-Zeiss” and substitute the word ‘Howard’. Third paragraph, line 1.—Insert after the word “with” the words “a magnification of’, making the sentence read: ‘Place the slide under the microscope and examine with a magnification of about 90 diameters, etc.” At the end of the paragraph add the following: “This area is of vital importance and may be obtained by adjusting the draw-tube to the proper length as determined by actual measurement of the field, a 16 mm. Zeiss apochromatic objective with a Zeiss X6 compensating ocular, or a Spencer 16 mm. apochromatic objective with a Spencer X10 compensating ocular, or their equivalents, being used to obtain the proper mag- nification.”” (5) 51, YEASTS AND SPORES.—TENTATIVE. Second paragraph, line 7.—Eliminate the last sentence of the paragraph and sub- stitute therefor the following: “‘Make the count with a magnification of about 180, to obtain which the following combinations, or their equivalents, should be employed: 8 mm. Zeiss apochromatic objective with X6 Zeiss compensating ocular, or an 8 mm. Spencer apochromatic objective with X10 Spencer compensating ocular with draw-tube not extended. (6) 52, BACTERIA.—TENTATIVE. Line 2.—Eliminate the sentence beginning ‘Use a magnification of about 500”, and substitute therefor the following: ‘Employ a magnification of about 500, which may be obtained by the use of an 8 mm. Zeiss apochromatic objective with an X18 Zeiss compensating ocular with draw-tube not extended, or an 8 mm. Spencer apochro- matic objective with an X20 Spencer compensating ocular with a tube-length of 190, or their equivalents.” Line 4.—Eliminate the sentence beginning ‘Because of the somewhat” and ending “being about 375.” 1920| REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 571 Mr. B. J. Howard (Bureau of Chemistry, Washington, D. C.) made the following recommendation: 49, 50, 51, 52, Micro-ANALysis OF TOMATO PULP, KETCHUP, PUREE AND SAUCE (PASTE). That the methods for the analysis of these products, 49, 50, 51 and 52, be trans- ferred to the chapter on canned vegetables. Adopted. Methods on spices and other condiments, as changed, adopted as a whole. XXV. CACAO PRODUCTS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 327-30, be changed as follows, and as changed be adopted as the official and tentative methods of the asso- ciation for the analysis of cacao products: CHANGE. 15, SUCROSE AND LACTOSE.—TENTATIVE. Line 9.—Before the word “liquid” insert the word “‘added”’. Methods on cacao products, as changed, adopted as a whole. XXVI. COFFEES. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 331-4, be adopted as the official and tentative methods of the association for the analysis of coffees. Adopted. XXVIII. TEA. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 335-7, be adopted as the official and tentative methods of the association for the analysis of tea. Adopted. 572 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 4 XXVIII. BAKING POWDERS AND THEIR INGREDIENTS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and published in the Association of Official Agricul- tural Chemists, Methods, 1916, 339-50, be changed as follows, and as changed be adopted as the official and tentative methods of the asso- ciation for the analysis of baking powders and baking chemicals: CHANGES. (1) XXVIII. BAKING POWDERS AND THEIR INGREDIENTS. Change the title of the chapter to “Baking Powders and Baking Chemicals.” (2) 3 (a), 50% potassium hydrozid solution. Eliminate the entire paragraph and substitute therefor the following: “(a) Potassium hydrorid solution—Dissolve 25 grams of potassium hydroxid in 50 cc. of water.” (6) LE BE 7/5. ts Make uniform use of parentheses with letters for designating parts of apparatus in the text. Methods on baking powders and their ingredients, as changed, adopted as a whole. XXIX. DRUGS. That the revised methods, as reported by the Committee on Editing Methods of Analysis and printed in the Association of Official Agricul- tural Chemists, Methods, 1916, 351-66, be changed as follows, and as changed be adopted as the tentative methods of the association for the analysis of drugs: CHANGES. (1) 1, PREPARATION OF SAMPLE AND SOLUTION.—TENTATIVE. (b) Line 8.—At the end of the sentence ending with the word “extraction’’ insert the following sentence: ‘‘Any caffein acetanilid mixture observable about the apex of the delivery tube of the separatory, edge of filter and tip of funnel should be very carefully recovered by judicious washing with chloroform, such washings being sub- sequently united with the main portion.” (C) Line 8.—After the word “in” insert the words “the aqueous”. (2) 2 (a), Standard bromid-bromate solution. Line 4.—Insert after the word “‘recrystallized” the words ‘“‘and dried’’. eRe ll me > —~I w 1920] REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 5 (3) 3, CAFFEIN.—TENTATIVE. First paragraph, last line.-—After the word “‘stand” insert the words “in the open”. Second paragraph, line 11.—After the word “‘liquid” insert the words “deep claret”. (4) 9, ANTIPYRIN.—TENTATIVE. Line 14.—After the word “‘above’’ insert the sentence: “Recover any crystalline product separating about the tip of the delivery tube and funnel and edge of filter by judicious washing with chioroform.” Insert in fine print between the first and second paragraphs the following: _ “The use of alcohol-free chloroform in connection with the halogenation of antipyrin is necessary in order to preclude the formation of iodoform. the presence of which in the composite residue “‘a’’ would vitiate the result.” (5) 10, caFFEIN.—TENTATIVE. Line 1 of (2).—Eliminate the word “Hydrolysis” and substitute therefor the words “Hydrolytic treatment”’. (6) 12, peTERMINATION. Line 6.—After the word “thoroughly” insert the words “by rotating the liquid’’. > Line 7.—Eliminate the word “shaking” and substitute therefor the words “rotating the liquid’. Line 14.—After the word “thoroughly” insert the words “by rotating the mixture’. Line 3 of (2).—After the word “‘transfer” eliminate the words “together with the filter” and substitute therefor the words “the precipitate together with the filter, likewise any particles of the precipitate remaining in the graduated flask”’. (7) 29, QUININ SULPHATE.—TENTATIVE. At the end of the paragraph add a separate paragraph in fine print as follows: “If the morphin sulphate present is contaminated with codein sulphate, the latter will be separated and weighed with the quinin.” (8) 30, MoRPHIN SULPHATE.—TENTATIVE. After the second paragraph add two additional paragraphs in fine print as follows: “Despite all precautions looking to the exclusion of impurities from the morphin as weighed, the amount of this substance thus determined will usually be greater than that found yolumetrically. In order to insure the greatest possible accuracy in volu- metric operations on alkaloidal residues like quinin, morphin and codein, it is suggested that, whenever possible. the strength of the standard acid used be checked by titration against the pure alkaloid under examination. “In the various operations involving fixation and subsequent liberation of morphin by means of fixed alkali and ammonium chlorid, the most careful attention should be paid to the manner of adding the reagents, since any undue excess of either might nullify the entire procedure. Any large excess of sodium hydroxid would naturally require for its reduction a correspondingly large amount of ammonium chlorid, the latter in turn yielding its equivalent of hydroxid, relatively large quantities of which through interaction with sodium chlorid tend to inhibit any permanent liberation of alkaloid and thus prevent complete extraction. Furthermore, ammonium chlorid in large amount operates retentively on the morphin in solution, due in part possibly to the formation of an alkaloidal hydrochlorid.”’ (9) 39, pREPARATION OF SOLUTIONS. (b) Line 2.—After the word “‘sample’’ add “(a)”. (C) Line 2.—After the word “‘sampie”™ add “(a)”. (d) Line 2.—After the word ‘“‘sample” add “(a)’’. 574. ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (10) 40, perermiation. Second paragraph, line 2.—After the word “sample” insert “39 (b)”. Second paragraph, line 3.—After the word “sample” insert “39 (C)”. Second paragraph, line 5.—After the expression “U. S. P. pepsin per cc.’’ insert teu Coys Second paragraph, line 8.—After the expression “pepsin per cc.” insert “38 (b)”. (11) 44, DISTILLATION.—TENTATIVE. Line 3.—After “200°C.” insert “in such a way that the mercury bulb shall be op- posite the side tube of the flask and the 175° mark below the cork.” Methods on drugs, as changed, adopted as a whole. Respectfully submitted, R. E. Doourrtie, Joun PuHILurrs STREET, W. A. WITHERS, A. F. SEEKER, G. W. Hoover, Committee on Editing Methods of Analysis. The report was adopted with a rising vote of thanks to the Committee on Editing Methods of Analysis and to the chairman of the committee for their splendid work and for the fine report they presented. As chairman of the Committee on Editing Methods of Analysis, Mr. Doolittle took occasion to thank not only the members of the association but outside chemists as well for the valuable service they rendered the committee. A motion was made, seconded and adopted that the Committee on Editing Methods of Analysis be continued and that it be instructed to incorporate in the methods the changes adopted at this meeting. Mr. Doolittle called attention to the fact that quite a number of tentative methods appear in the revised methods, which the Committee on Editing Methods of Analysis feel have been before the association long enough either to be made official or rejected. The committee, however, did not know how to handle the matter and so Mr. Doolittle, as chair- man of the committee, asked the association for instructions. A motion was adopted that this matter be left to the discretion of the Committee on Editing Methods of Analysis. 1920| REPORT OF COMMITTEE ON EDITING METHODS OF ANALYSIS 575 Dr. P. F. Trowbridge (Agricultural Experiment Station, Agricultural College, N. Dak.) raised the question of the advisability of designating methods by the names of authors. Considerable discussion ensued and, while the association approved the general principle of eliminating such names wherever possible, the following motion introduced by Dr. Trow- bridge was adopted: That the Committee on Editing Methods of Analysis be given authority to include or omit the names of authors in connection with the titles of methods, as in their judgment seems best. The meeting adjourned at 12.30 P. M. to reconvene at 1.30 P. M. 1915 Nov. 17 1916 April 15 Noy. 16 THIRD DAY. WEDNESDAY—AFTERNOON SESSION. REPORT OF SECRETARY-TREASURER FOR By C. L. AtsperG (Bureau of Chemistry, GENERAL RECEIPTS. Banksbalarice.~ 55 e BD - ae Es ey ee eee $181.04 Refund on letterheads ordered from Charles G. Stott and Company 75 1914-15 dues from 2 States (Alabama and Kansas)_____-_-___--_-_ 4.00 Dues for the year 1915-16 from 84 Federal, State, and municipal organizations: = 3222-68 ee eee 168.00 1916—17 dues} fromyPorto Rico 2! 3 2 eee $355.79 JOURNAL RECEIPTS. Subscription from Kansas State Board of Health_______-___-___-- $4.00 Subscription from Pittsburgh Bureau of Food Inspection (F. C. Buckmaster); 2)... sos oe See eee Subscription from Laboratory Inland Revenue, Vancouver, B. C. G: A“ Dawson)/s2. 22. oe ee a SS eee 4.00 Subscription from Tennessee State Department of Agriculture (J: |W..;Sample)o22 2. 23 eee 4.00 Deficit, paid from Secretary-Treasurer account- ~~ ~~ ..-.-.------- 14.82 $30.82 or i | for) 1920] ALSBERG: REPORT OF SECRETARY-TREASURER 577 THE YEAR ENDING NOVEMBER 22, 1916. Washington, D. C.), Secretary-Treasurer. ACCOUNT. DISBURSEMENTS, 1915 Noy. 17 Telephone calls, Raleigh Hotel, 1915 meeting____________________ $1.10 INOVeNU eee lips witaleightiotel: (25002) Seer IG Eis Te Ey ee ae 2.50 1916 Janel On sbostionice box rent, (check, No..37)-_-=——_- -=--<------ 1.00 Feb. 3 Printing circulars and bills (check, No. 39)_______________._-__-___- 17.75 Bieb wae mem ostage: (CheckyiNo: 41) 22 4s es ee 5.00 Heb wetigee bostape: (check INO:143) 22-225. [=o See 3.00 Aprile ee bostroitice box.rent) (check No: 45) 29252 _ 2525" 52 2 ie ee 1.00 Aprilae4 mmbostages (CheCKsINO:(46))6 2 52-2 ee ae oe ee 2.00 Apidae sammbetverheads) (CheckwINO:,48) = =o ae ee ee eee 7.50 April18 1000 special request envelopes (check No. 49)___-_______________- 22.00 Aprmiltsambostagey(check#No50) 5.22. = a ee 2.00 July IRostronicesboxsrenth( ChECKs NO iO 1) see eee te ne ee 1.00 Nuge2+eooorenvelopes (check*No:) 55) 22-225. = Sia se eee 75 Depiy pm ostolice/ box: rent) (check. No: 57) 22-2222. 1.00 Oct Ome bostaper CHECKSING 7 08))= =. —- == ane eee es oe ee 4.25 Oct. 30 One ream typewriter paper (check No. 59)___._.-__------------- 1.80 Nov. 15 Printing 800 announcements, 1916 meeting (check No. 60) -___---- 23.25 Wovesko moo) bad wess(cGhecks NO: iOL)E 2. - See ee ee 23.00 Donmrnalaccaur yee eee eae 2 SE as Se See ee 14.82 iNoven Gee bankabalances=-) 22) a te ee $275.30 esstGhecks OU tte y= o n ee see oe Oe ne ee ee 54.23 221.07 $355.79 ACCOUNT. DISBURSEMENTS. 1915 “ Dec. 4 Williams & Wilkins Co. (Subscription Kansas State Board of Health, ChecksINosoG) 2-22 hes = hse pe ae ee es ee ee ee $4.00. 1916 Feb. 2 Williams & Wilkins Co. (Subscription F. C. Buckmaster, check INO) BI) ek te oe See ee ee Se er ee oe ee eee 4.00 Feb. 8 Williams & Wilkins Co. (Subscription J. A. Dawson, check No. 42) 4.00 Feb. 26 Williams & Wilkins Co. (Subscription J. W. Sample, check No. 44) 4.00 April 6 Expressage on methods manuscript (check No. 47)___------------ -30 Au oO ME EOStAPE] CHECKING S02) =2— ae ee eee 5.00 ip some lclerram (Cheeks NOs 03) cas 08 nea ne eee ee Se eee AL Angie Melegramy (check No./54)---- == 553-25. 25 5. 322 222 2225 2- se eke stile Sept. 14 Clerical work (Miss Ferriter, check No. 56)_------------------- 8.34 $30.82 The undersigned committee has examined the above report and finds it correct. Joun Puiiuires STREET, Cuas. B. LipMan, B. B. Ross, Auditing Committee. Approved. 578 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 REPORT ON THE JOURNAL. By C. L. Atssperc (Bureau of Chemistry, Washington, D. C.), Chairman, Board of Editors. It is possible to report only on Volume I, the last number of which was issued in June. There were over nine hundred subscriptions to that volume, and a deficit of only two hundred and seventy-five dollars, as shown in the statement received from the publishers. The statement from the publishers was not quite clear in certain respects, so an expert accountant is at present auditing the books. When Volume II began, we had a loss of one hundred and fifty to two hundred subscribers from failure to renew. Since that time, there has been a steady increase in the number of subscriptions. We receive from ten to twenty-five new subscriptions a month without special effort or solicitation. The present subscription list is approximately eight hundred. The publishers antici- pate that by spring or summer the list of subscribers will reach one thousand or twelve hundred. The indications are that probably there will be no deficit for Volume IT, so I think we can feel that the success of The Journal is now assured. [ think that a deficit of two hundred and seventy-five dollars on the first volume of a scientific journal of this nature is a good showing for the present year. There have been a good many subscriptions received from all over the world, some from South America, South Africa, Australia, and from each of the British Colonies. A number of subscriptions have also been received from France, England and Russia, and I dare say that this num- ber will be increased considerably at the end of the war. In reference to the methods, you will be interested to learn that the publisher is ready to reprint the methods as a separate volume. At present it is impossible to fix a definite price for the book, since we do not know how much it is going to cost to make these revisions and to incorporate into the revised methods the action taken at this meeting. Of course, I suppose it would be desirable to include everything done at this meeting. It is really very difficult to say definitely what it is going to cost, but we hope to be able to secure a very reasonable rate for the members of the association. The editors of The Journal have been endeavoring to dispose of the hitherto unpublished proceedings of the association and the revision of the methods. Judging from present prospects, we hope to have all the proceedings up to date by midwinter or early spring, including the proceedings of this meeting. If it were not for the portion of the 1915 proceedings still unpublished, we should be able to print the proceedings of this meeting much more rapidly than will be the case. At any rate, by midwinter or spring of the coming year we hope to be thoroughly 1920] ALSBERG: REPORT ON THE JOURNAL 579 up to date, when The Journal will be in a position to accept scientific communications of a research nature, apart from the proceedings. The editors of The Journal hope that, when The Journal begins to accept such communications, it may devote its pages to publishing research papers of a high grade, apart from reports of referees. It is the intention of the editors of The Journal, unless otherwise instructed by this association, to give preference to papers dealing with methods of analysis. It is not their intention, however, unless you wish otherwise, to reject good work in general agricultural chemistry when such work is presented for publication in The Journal. It is merely our intention to give preference, so far as space may require a discrimination, to analytical papers rather than to papers of a general chemical nature. Few editorial changes have been made in the reports of referees, except where there were obvious errors, misstatements or unnecessary para- graphs, because the Board of Editors was uncertain as to the extent of its authority in such matters. The editors of The Journal have accordingly asked the Executive Committee for instructions, and in com- pliance with this request the committee has drawn up the following resolution: Resolved, That the Board of Editors of the Journal of the Association of Official Agri- cultural Chemists be authorized to edit the reports of referees, and if any material changes are deemed advisable, that such proposed changes be submitted to the Chair- man of the Committee on Recommendations of Referees and Revision of Methods, and be referred to the proper Subcommittee A, B, or C for approval. The manuscript, as edited, shall be submitted to the referee for approval before publication. The Board of Editors is authorized to prepare a type form for the reports of referees which shall be followed by the referee wherever practicable, so that uniformity in the work of referees may be promoted and the work of editing such reports may be facilitated. The Board of Editors would welcome criticisms, advice, and recom- mendations, formal or informal, from the floor now or in writing at any time. The report, together with the resolution contained therein, was unanimously adopted by the association. Dr. CG. L. Alsberg, on behalf of the Executive Committee, recom- mended that the distribution of referees be revised. The essential features of the revision involved the grouping of the subjects more logically, the dropping of the general subject of food adulteration, and the appointment of a referee on micro-analytical methods. The recom- mendations of the committee were adopted. Dr. P. F. Trowbridge made the following recommendations: 580 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 (1) That the referee on meat and meat products make a special study of starch, glycogen, and the two nitrate methods. Adopted. (2) That an associate referee on meat extracts be appointed and that he be instructed to study during the next year the glycerol method, the sugar method, and the acid alcohol-soluble nitrogen method. Adopted. REPORT OF COMMITTEE TO COOPERATE WITH OTHER COMMITTEES ON FOOD DEFINITIONS AND STANDARDS. Your committee desires to submit the following report of the activities ~ of the joint committee on food definitions and standards during the past association year. Three meetings of the joint committee have been held during this year. Of these, the first of the year (the seventh conference of the com- mittee since its organization) was held in Washington, D. C., January 17 to 20, 1916; the second, in Detroit, August 5 to 10, 1916; the third, in Washington, D. C., November 16 to 18, 1916. These meetings were in part given up to hearings which were either granted upon request or were held upon published announcement at the committee’s initiative. Other hearings authorized by the joint com- mittee were held by subcommittees. Among the subjects of these hearings were: The grading of milk and cream, condensed milk prod- ucts, dried milk, homogenized and emulsified milk products, cheese, evaporated apples, spices, citrus fruits and baking powders. Conferences also upon the subject of grades of canned foods have been held with various manufacturing and distributing organizations. Various subcommittees have under consideration important groups of food products, some of which have been preliminarily considered by the joint committee, with respect to which the work is not yet sufficiently advanced to warrant announcement at this time. The following schedules have been adopted during the year by the joint committee for recommendation to the Association of American Dairy, Food and Drug Officials, the Association of Official Agricultural Chemists and the Secretary of Agriculture for their approval, and have been approved by the association first named, at its meeting of August 10, 1916. PRINCIPLES OF STANDARDIZATION. The general considerations which have guided the joint committee on definitions and standards in preparing definitions and standards for food products are the following: ) Presented by William Frear. 1920] REPORT ON FOOD DEFINITIONS AND STANDARDS 581 (1) The definitions are framed so as to include those facts of material, quality, origin and mode of preparation that are essential to distinguish the food named. These definitions may or may not be accompanied by specifications of limits of physical quality and chemical composition characteristic of the food defined. Foods vary in composition with differences in season and soil, and because of varia- tions in manufacturing operations, and they may be associated with small amounts of foreign substances, owing to imperfect conditions of production. The specifcations of limits of physical quality and chemical composition are so drawn as to "provide for variations due to such causes and of degree commonly accepted as reasonable. (2) The definitions are so framed as to exclude from the articles defined substances not included in the definitions. (3) A term defined in any of the several schedules has the same meaning wherever it is used in any schedule. (4) Thenamesof food products herein defined preferably agree with existing American usage as known to the consumer. MILK PRODUCTS. Definitions adopted by the joint commitiee on definitions and standards, August 6, 1916: Sweetened condensed milk is the product resulting from the evaporation of a con- siderable portion of the water from milk to which sugar (sucrose) has been added. It contains, all tolerances being allowed for, not less than twenty-eight per cent (28%) of total milk solids, and not Jess than eight per cent (S%) of milk fat. Condensed skimmed milk, evaporated skimmed milk, concentrated skimmed milk, is the product resulting from the evaporation of a considerable portion of the water from skimmed milk, and contains, all tolerances being allowed for, not less than twenty per cent (20%) of milk solids. Sweetened condensed skimmed milk, sweetened evaporated skimmed milk, sweetened con- centrated skimmed milk, is the product resulting from the evaporation of a considerable portion of the water from skimmed milk to which sugar (sucrose) has been added. It contains, all tolerances being allowed for, not less than twenty-eight per cent (28%) of milk solids. Dried milk is the product resulting from the removal of water from milk, and con- tains, all tolerances being allowed for, not less than twenty-six per cent (26%) of milk fat, and not more than five per cent (5%) of moisture. Dried skimmed milk is the product resulting from the removal of water from skimmed milk and contains, all tolerances being allowed for, not more than five per cent (5%) of moisture. Malted milk is the product made by combining whole milk with the liquid separated from a mash of ground barley malt and wheat flour, with or without the addition of sodium chlorid, sodium bicarbonate and potassium bicarbonate, in such manner as to secure the full enzymic action of the malt extract, and by removing water. The result- ing product contains not less than seven and one-half per cent (7.5%) of butter fat and not more than three and one-half per cent (3.59) of moisture. EDIBLE VEGETABLE FATS AND OILS. Definitions adopted by the joint committee on definitions and standards, November 18, 1916: Edible fats and edible oils are such glycerides of the fatty acids as are recognized to be wholesome foods. They are dry and sweet in flavor and odor. 582 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 Cacao butter, cocoa butter, is the edible fat obtained from sound cacao beans (Theobroma cacao L..), either before or after roasting. Coconut oil, copra oil, is the edible oil obtained from the kernels of the coconut (Cocos nuctfera L. or Cocos butyracea L.). Cochin oil is coconut oil prepared in Cochin (Malabar). Ceylon oil is coconut oil prepared in Ceylon. Corn oil, maize oil, is the edible oil obtained from the germ of Indian corn, maize (Zea mays L.). Cottonseed oil is the edible oil obtained from the seed of the cotton plant (Gossypium herbaceum, L.) or from the seed of other species of Gossypium. Olive oil, sweet oil, is the edible oil obtained from the sound, mature fruit of the olive tree (Olea europaea L.). Palm kernel oil is the edible oil obtained from the kernels of the fruit of the palm tree (Hlaeis guineensis L. or Elaeis melanococca Girt.). Peanut oil, arachis oil, earthnut oil, is the edible oil obtained from the peanut (Arachis hy pogea L.). Poppy seed oil is the edible oil obtained from the seeds of the poppy (Papaver somni- ferum L.). Rape seed oil, rape oil, colza oil, is the edible oil obtained from the seed of the rape plant (Brassica napus L.), or from the seed of closely related Brassica species, which yields oils similar in composition and character to the oil obtained from the seed of Brassica napus L. Soy bean oil, soy oil, soja oil, is the edible oil obtained from the seed of the soy bean plant (Glycine soja L., Soja hispida, Sieb et Zuce., Soja max. (L.) Piper). Sesame oil, gingili oil, teel oil, benne oil, is the edible oil obtained from the seed of the sesame plant (Sesamum indicum, De Candolle, Sesamum radiatum, Schum and Thonn, Sesamum orientale L.). Sunflower oil is the edible oil obtained from the seed of the sunflower (Helianthus annuus L.). EVAPORATED APPLES. Definition adopted by the joint committee on definitions and stand- ards, August 7, 1916: Evaporated apples are evaporated fruit made from peeled, cored and sliced apples and contain not more than twenty-four per cent (24%) of moisture. Pending the official adoption by this association of perfected methods for estimating moisture in evaporated apples, the following trade method shall be employed: Dry a representative unminced sample for 4 hours at the temperature of boiling water and determine the loss in weight. In addition to the foregoing, partial schedules for “Soda Water Flavors” and for “Soda, Soda Water’? were recommended to and ap- proved by the Association of American Dairy, Food and Drug Officials in August, 1916. Owing, however, to the publication of the results of an investigation which may require the modification of a fundamental conclusion involving these schedules, the joint committee has deemed it wise to withhold them from presentation at this time, so that they may be further considered before the approving acts of the several authorities concerned shall have been completed. 1920) REPORT ON FOOD DEFINITIONS AND STANDARDS 583 The joint committee has prepared for your present action two groups of definitions and standards, in addition to those above given. They deal with macaroni and related products and with baking powders. MACARONI, SPAGHETTI, VERMICELLI. FLOUR MACARONI, FLOUR SPAGHETTI, FLOUR VERMICELLI. Definitions and standards adopted by the joint committee on defini- tions and standards, November 18, 1916: Macaroni, spaghetti. vermicelli are dried pastes made of the semolina of hard wheat. They contain not more than thirteen and one-half per cent (13.5%) of moisture. Flour macaroni, flour spaghetit, flour vermicelli are dried pastes made of flour or of a mixture of flour and semolina. They contain not more than thirteen and one-half per cent (13.5%) of moisture. BAKING POWDER. Definition adopted by the joint committee on definitions and stand- ards, November 18, 1916: Baking powder is the leavening agent produced by the mixing of an acid reacting material! and sodium bicarbonate, with or without starch or flour. It yields not less than twelve per cent (12%) of available carbon dioxid. The acid reacting materials in baking powder are: (1) Tartaric acid or its acid salts; (2) acid salts of phosphoric acid; (3) compounds of aluminium; or (4), any combination in substantial proportions of the foregoing. COMMENTS. The milk products schedule is, with the exception of the definition and standard for sweetened condensed milk, entirely new matter. The definition for sweetened condensed milk, though modified in phrasing, is substantially a reaffirmation of that previously proclaimed’, with a very slight increase of the fat minimum. The schedule for edible vegetable fats and oils represents a revision of the corresponding schedule of Circular 19. The revision differs from the former schedule in having a group definition comprehending a state- ment of the chemical character of these products and, in addition, of those qualities that determine edibility; in the omission from the specific definitions of all matters covered by the group definition; in the dropping of separate standards for cold-pressed and virgin oils; in the omission, for the present, of statements of chemical limits, those of 1The announcement of the amount of calcium sulphate which reacts as an acid reacting material in baking powder is reserved pending further investigation. Bis es ee: he Baking powder materials should be a- free from metallic impurities as it is feasible for a manu- facturer to make them. The announcement of the limits for arsenic, lead, zinc and fluorids is reserved pending further investigation. 2U.S. Dept. Agr., Office of the Secretary, Circ. 19. 584 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 Circular 19 having been found confusing because of the general over- lapping of the so-called ‘‘constants” of different oils, and the wide varia- tion of these characters in individual oils without corresponding, clearly defined alteration of their food values; and, finally, in the addition of a definition for soy bean oil. The definition and standard for evaporated apples is a revision of the corresponding definition and standard of Circular 19. It differs there- from in the inclusion of “slicing” as a process in the manufacture, in a reduction of the moisture maximum, and in a fuller description of the commercial method of moisture determination for this product. Permit us, in this connection, to note the need for a special study by this asso- ciation of the problem of evolving a more exact and yet convenient method for this determination. The schedule for macaroni and related products proposes what is deemed the best solution of the difficulties arising from the manufacture and sale of alimentary pastes made of wheat flour, or of mixtures thereof with semolina, in the form of macaroni, vermicelli and spaghetti normally made exclusively from the semolina of hard wheat. The definition and standard for baking powder deals fundamentally with a group of products that have, from time to time, been the subjects of much discussion, and which present many phases of composition. The definition with its accompanying declarations provides for the use, as ingredients of baking powder, of all materials generally employed as such ingredients. It suggests the suitable subordinate classification according to the acid or acid-reacting substance or substances used; and, finally, it leaves the way open to the introduction, under suitable label declarations, of wholesome food materials which may for any reason seem desirable, providing such introduction does not in any way contravene the Federal Food and Drugs Act, and does not reduce the yield of available carbon dioxid below the accompanying specification. This specification is in accordance with the producers’ recommendations and not below the limit fixed by those State laws in which a limit for this leavening reaction product has been given. Respectfully submitted, WitiiAM FREAR, Junius Hortvet, Joun PHILLIPS STREET, Committee to Cooperate with Other Committees on Food Definitions and Standards. Approved. 1920) AVAILABILITY OF PHOSPHORIC ACID IN BASIC SLAG 585 REPORT OF COMMITTEE ON AVAILABILITY OF PHOSPHORIC ACID IN BASIC SLAG". Your committee desires to submit the following report of progress made during the past association year. The Rhode Island Agricultural Experiment Station has completed the work it has been conducting in cooperation with your committee and has submitted its final report. This includes pot and field experiments with millet and rape. The year 1916 was devoted to observing the after- effects in the field of the phosphoric applications made during 1915. The Hawaii, Texas, New York (Ithaca), and North Dakota Agricul- tural Experiment Stations have submitted no further data than that referred to in the last report of your committee?. The Massachusetts Agricultural Experiment Station reports that three series of pot experiments with rape and millet have been completed and that one season’s growth of rape with the different phosphates has been attained in the field. The results, however, could not be reported prior to this meeting because of contemplated chemical work on the crops. At the Pennsylvania Agricultural Experiment Station clover and timothy were grown on the field plots in 1916 without any phosphatic application since 1913. It is probable that a new application of phos- phatic materials will be made in the near future. A report from the Illinois Agricultural Experiment Station includes results of pot experiments conducted during 1915 and 1916 with clover, soy beans, wheat, millet and rape. Data obtained with phosphates other than those furnished by the committee are also included. Owing to inequalities still existing in the fields selected for the experi- ments at the North Carolina and Virginia Agricultural Experiment Stations, these stations postponed the application of phosphates. Pot experiments have been started at the North Carolina Agricultural Experiment Station with prospect of a final report in 1917. Your committee feels that the referee on phosphoric acid should be instructed to continue the study of methods for determining the availa- bility of phosphoric acid in slag, and chemical matters pertaining thereto, so that the association may have at hand the data necessary to aid it in the adoption of an availability method as soon as the results of the vegetation experiments are obtained. Sufficient reports are already in the hands of your committee to be of service to the referee on phosphoric acid in his chemical investigations. It seems unnecessary to wait until all of the vegetation results are at hand before tentative methods of analysis are submitted to the association. 1 Presented by P. F. Trowbridge. 2 J. Assoc. Official Agr. Chemists, 1917, 3: 104. 586 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 It is therefore recommended by your committee that this association instruct its referee on phosphoric acid to give prominent attention to the question of methods of determining available phosphoric acid in slag, the chemical ingredients influencing the same, and the bibliography on the subject. C. B. W1LLIAMs, H. D. Haskins, B. L. Hartwe tt, C. G. Hopkins, J. A. BizzEL1, Committee on Availability of Phosphoric Acid in Basic Slag. Adopted. REPORT OF COMMITTEE ON AMENDMENTS TO THE CON- STITUTION AND BY-LAWS. By B. B. Ross (Polytechnic Institute, Auburn, Ala.), Chairman. Your committee recommends the adoption of the Constitution and By-laws in the following amended form: CONSTITUTION OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS. ARTICLE I. This association shall be known as the AssocIATION OF OFFICIAL AGRICULTURAL CHEmists OF NortH AMERICA. The objects of the asso- ciation shall be (1) to secure uniformity and accuracy in the methods, results, and modes of statement of analysis of fertilizers, soils, cattle foods, dairy products, human foods, medicinal plants, drugs, and other materials connected with agricultural industry; (2) to afford opportunity for the discussion of matters of interest to agricultural chemists. ‘ ARTICLE Il. Analytical chemists connected with the United States Department of Agriculture, or with any State, provincial, or national agricultural experi- ment station or agricultural college, or with any State, provincial, or national institution or body in North America charged with official control of the materials named in Article I, shall alone be eligible ex officio to activemembership. Analytical chemists connected with municipal lab- oratories charged with control of any of the materials or subjects named in Article I shall be eligible ex officio to associate membership. Active members of the association who lose their right to such membership by re- tiring from positions indicated above as requisite for eligibility to active membership may become eligible ex officio to associate membership. Persons may be elected to honorary membership by the two-thirds vote of those present at any regular meeting of the association. ~I 1920] ROSS: AMENDMENTS TO CONSTITUTION AND BY-LAWS 58 ARTICLE Il. The officers of the association shall consist of a President, a Vice- President, and a Secretary, who shall also act as Treasurer, and these officers shall be elected annually from active members. The duties of said officers shall be those that generally pertain to such positions. These officers, together with two other active members, to be elected by the association, shall constitute the Executive Committee. The special duties of the officers of the association shall be further defined, when necessary, by the Exeeutive Committee. There shall be appointed by the President, on the recommendation of the Executive Committee, a committee of nine members, which shall be designated as a Committee on Recommendations of Referees, one-third of the membership of which shall be appointed at intervals of two years to serve six years. The chairman of the committee shall be elected by the members of the com- mittee from among their own number. The chairman shall divide the nine members into subcommittees (A, B, and C) and shall assign to each subcommittee the reports and subjects which it shall consider. At the annual meeting there shall be appointed by the President, upon the recommendation of the Committee on Recommendations of Referees, from among the members! of the association, a referee and associate referees for each of the subjects to be considered by the association. It shall be the duty of these referees to prepare and distribute samples and standard reagents to members of the association and others desiring the same, to furnish blanks for tabulating analyses, and to present at the annual meeting the results of the work done, discussion thereof, and recommendations for methods to be based thereon. ARTICLE IV. The annual meeting of the association shall be held at such place as shall be decided by the association, and at such time as shall be decided by the Executive Committee, and announcement thereof shall be made at least three months prior to the time of said meeting. Special meet- ings may be called by the Executive Committee when in its judgment it shall be necessary. ARTICLE V. All proposed changes or amendments to this constitution shall be presented in writing and read in full to the association not later than the first day of the regular annual meeting, shall be referred to the 1 When used without any qualifications, construed by the association to mean either active or associa te members. 588 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 Executive Committee, and after a report from this committee may be adopted on the succeeding day by a vote of three-fourths of the active members present. Adopted. BY-LAWS OF THE ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS. (1) Any amendment to these by-laws or changes therein may be proposed and adopted in the same manner as amendments to the con- stitution, but only a two-thirds vote of the active members present shall be required for their adoption. (2) These by-laws or any portion of them may be suspended at any regular meeting of the association without previous notice by a vote of three-fourths of the active members present. (3) Only such colleges, experiment stations, bureaus, boards, or other institutions whose members are active members of this association shall be entitled to enter motions and vote. (4) On general questions before the whole association, each college, experiment station, etc., as qualified above, shall be entitled to one vote only. In voting upon questions involving methods of analysis, defini- tions, nomenclature and laws or regulations relating to materials men- tioned in Article I of the Constitution, each of the said institutions shall be entitled to vote only upon questions relating to those materials over which said institution exercises official control. (5) A method shall not be adopted as official or an official method be amended until such method or amendment has been recommended as official for at least two annual meetings by the appropriate referee. (6) No changes shall be made in the methods of analysis used in official inspection until an opportunity shall have been given all active members having charge of the particular inspection affected to test the proposed changes. (7) A method shall not be adopted as tentative or a tentative method amended until such method or amendment has been reported by the appropriate referee and published in the proceedings of the association. (8) When any officer, referee, or associate referee ceases to be eligible for membership in the association, his office shall be considered vacant and a successor may be appointed by the Executive Committee to con- tinue in office until the next following regular meeting. The Executive Committee shall also have authority to fill vacancies occurring in any other manner. (9) Chemists and others interested in the objects of the association may attend its meetings, take part in its discussions, and present papers, if permission is secured from the Executive Committee. 1920) FREAR: REPORT OF COMMITTEE ON RESOLUTIONS 589 (10) Each college, experiment station, bureau, board or other insti- tution entitled to representation in the association shall contribute annually $5.00 prior to the first of January following the regular annual meetings. Adopted. A motion was made, seconded and adopted that the Executive Com- mittee give careful consideration to definitions for “‘official method” and “tentative method”’, and formulate definitions to be presented next year, and that at the present time the words be without definition. REPORT OF COMMITTEE ON NOMINATIONS. By C. C. McDonne tu (Bureau of Chemistry, Washington, D. C.), Chairman. The committee submitted the following nominations for officers for the year ending November, 1917: President, J. K. Haywood of Wash- ington, D. C.; vice-president, P. F. Trowbridge of North Dakota; secretary-treasurer, C. L. Alsberg of Washington, D. C.; additional members of the executive committee, B. B. Ross of Alabama, and H. C. Lythgoe of Massachusetts. The secretary was instructed to cast the unanimous ballot of the asso- ciation for these officers. REPORT OF COMMITTEE ON RESOLUTIONS'. By Wituram Frear (Agricultural Experiment Station, State College, Pa.), Chairman. We have lost from our number, through death since the last meeting, four members: Robert James Davidson, on December 19, 1915; Eugene Woldemar Hilgard, on January 8, 1916; George Edward Patrick, on March 22, 1916; and Thomas Cuthbert Trescot, on April 14, 1916. We mourn the loss of these tried and true associates, and desire to record in some permanent way our appreciation of them. Owing to the long interval between the time of Professor Davidson’s death and this: meeting, the officers of this association thought best to draft and present to his widow the following memorial: 1 Presented by W. A. Withers. 590 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS | Vol. III, No. 4 Wuepreas Providence has removed from membership in this associa- tion our honored former president and coworker, Robert James David- son of Virginia: Therefore be it Resolved, That in the death of Professor Davidson the Association of Official Agri- cultural Chemists has lost one of its staunchest and most trusted workers, and its members, an associate admired for his thoroughness and steadfastness of judgment, and beloved for his modesty and rare kindliness of spirit. He shot no barbed shafts of wit; he wielded no double-edged sword of satire; nor opened his lips to utter a word of dis- paragement of others; but wherever he was, there gathered a group drawn by the lode- stone of his character, the smile of friendship upon his often pain-worn face, his sym- pathetic heart, and his keen discernment and high appreciation of what was good and true in others. Through a quarter of a century he has been constantly contributing to the work of the association, rendering service of especial merit in relation to the methods of analysis for insecticides and fungicides, and has acted as the referee on these methods and upon those for determining nitrogen, and also, for some years, as chairman of one of the three most responsible committees on recommendations of referees and revision of methods. In 1903, in recognition of his labors for the association and of his high standing in the chemical profession, he was chosen president of the association. In 1907 he was made the delegate of the association to report to the International Congress of Applied Chemistry at the London meeting, the association’s judgment concerning the unifica- tion of terms to be used in reporting analytical results. His contributions to the Ameri- can literature of agricultural chemistry include some of its best work upon the subjects of the chemistry of tobacco, and of fruits and fruit juices. Resolved, further, That this association express to the Virginia Agricultural and Mechanical College and Agricultural Experiment Station its regret at the loss of so able and useful an officer, teacher, and investigator, and to Mrs. Davidson its sympathy in her sorrow. January 19, 1916. Your committee recommends the adoption of the following resolution: Resolved, That the Association of Official Agricultural Chemists reaflirms the memorial to Robert James Davidson drafted by its officers, and directs that it be spread upon the minutes, and that the secretary send to Mrs. Davidson a copy of this resolution. Although Professor Hilgard, owing to the burden of administrative duties and his remoteness from our meeting place, and, in later years, because of his failing health and advancing age, was unable to attend many of the sessions, he was nevertheless an ex officio member, and maintained to his death a warm interest in such of our work as pertained to his special field of study. Your committee recommends the adoption of the following resolution as a brief memorial of his service: Resolved, That in the death of Professor Eugene Woldemar Hilgard, Ph. D., LL. D., America has lost her leading pioneer in the domain of soil chemistry. To his long service, comprehensive labors, keen insight, sense of proportion, stability of judgment, 1920) FREAR: REPORT OF COMMITTEE ON RESOLUTIONS 591 clearness of expression and wealth of inspiration, American agricultural chemists owe a lasting debt. The Association of Official Agricultural Chemists directs that this reso- lution of appreciation be placed upon its permanent records, and that its secretary transmit a copy thereof through the University of California to Dr. Hilgard’s family The adoption of the following resolution, memorial to Professor Patrick, is recommended: Resolved, That in the death of Professor George Edward Patrick, from 1873 to 1874, instructor in chemistry in Cornell University; from 1874 to 1883, professor of chemistry in the University of Kansas; from 1888 to 1895, chemist of the Iowa Agricultural Experiment Station; from 1895 to 1901, assistant chemist; and from 1901 to his death, Chief of the Dairy Laboratory of the Bureau of Chemistry; author of the Patrick method for milk fat determination; deviser of the copper distillation flask for the Kjeldahl nitrogen method; sometime referee of this association; skilled craftsman in the application of analytical methods to the examination of dairy products, this associa- tion has lost a valued collaborator and a comrade, whose sterling worth we have tested through twenty years of acquaintance. Resolved, further, That this memorial be spread upon the minutes, and a copy trans- mitted to his nearest of kin Your committee recommends the adoption also of the following reso- lution, memorial to Mr. Thomas Cuthbert Trescot: Resolved, That in the death of Thomas Cuthbert Trescot, assistant chemist in the Bureau of Chemistry since 1884, for many years Chief of the Nitrogen Laboratory of the Bureau of Chemistry, sometime referee of this association, expert analyst of nitrogenous compounds, this association has lost a valued collaborator and a highly esteemed friend. Resolved, further, That this resolution be spread upon the minutes, and a copy thereof transmitted to his widow Your committee further recommends that the Board of Editors be instructed to have prepared and to print in The Journal of this associa- tion suitable biographical sketches of these deceased members. Adopted by rising vote. REEL LT LTE IE OE | ESAT TE ROBERT JAMES DAVIDSON. Robert James Davidson, Dean of the Department of Applied Science and Professor of Chemistry at the Virginia Polytechnic Institute, died very suddenly at his home in Blacksburg, Virginia, December 19, 1915. He was born in Armagh, Ireland, April 3, 1862, of Scotch-Irish parents. He was only an infant when his father died and the family moved to Manchester, England, where he received his early education. When he was about sixteen years of age he went to Georgetown, 5. C., and made his home with an uncle. In 1882 he matriculated at the South Carolina College at Columbia, where he completed a four year course in 592 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 three years, doing extra work on the campus to maintain himself while at college. He received the degree of Bachelor of Science in 1885, and returned the following session as tutor in chemistry and as secretary of the faculty, pursuing at the same time advanced studies in chemistry. He received his Master of Arts degree in 1887. In 1888 he was appointed Assistant Professor of Chemistry and Assistant Chemist of the South Carolina Experiment Station. In 1891 he was elected Professor of Chem- istry in the Virginia Agricultural and Mechanical College at Blacksburg, Virginia, and also Chemist of the Virginia Agricultural Experiment Sta- tion. He was married on May 2, 1892 to Miss Anna McBryde. His wife and two daughters survive him. In 1904 he was elected to the position of Dean of the Department of Applied Science in the Virginia Polytechnic Institute, where he served with great efficiency until the day of his death. His duties as an instructor made heavy demands upon his time and energy, yet he kept in touch with the progress of science. He was a Fellow of the American Association for the Advancement of Science; a member of the American Chemical Society; and a member of the Asso- ciation of Official Agricultural Chemists, attending the meetings of these societies whenever possible. He contributed articles along agricultural chemical lines to the publications of the Virginia Agricultural Experi- ment Station. He always looked forward with pleasure to the meetings of the Association of Official Agricultural Chemists and attended nearly all of them for a period of over twenty years. He was always active in all the work of the association, serving as referee on several subjects and cooperating in others. In 1903 he served the association as its president. Professor Davidson was endowed with exceptional ability as a teacher, and his life and character was a source of inspiration and encouragement to all who came into contact with him. W. B. Evert. EUGENE WOLDEMAR HILGARD. AN APPRECIATION. Eugene Woldemar Hilgard was born at Zweibrucken, Rhenish Bavaria, January 5, 1833. His father came to America, with his family, three years later and settled on a farm in Illinois, and, owing to the crude nature of the public schools there, undertook personally the training of his son. Thus Eugene was ready for the university at sixteen, and was sent to Germany. He studied at Heidelberg, receiving there the doctor's degree in 1853, at the age of twenty. That degree was reissued to him 1920) LIPMAN: OBITUARY ON EUGENE WOLDEMAR HILGARD 593 by Heidelberg in 1903 as a “Golden Degree” in recognition of his dis- tinguished services to science in fifty years. After taking his degree, Dr. Hilgard went to Spain, where he met Miss J: Alexandrino Bello, whom he married in 1860. In 1855 he returned to America, where he did geological work in Mississippi until 1858, when he was appointed State Mineralogist of that State. During the Civil War, Hilgard was a scientific adviser to the Confederacy. In 1866 he was made Professor of Chemistry in the University of Missis- sippi, and later Professor of Zoology, Geology and Botany. In 1872 he was made Professor of Geology and Natural History in the University of Michigan. That title amused him much then and in after years, as I remember his gleeful chuckle in telling about it. In 1874 he was called to the University of California, where he remained until his death. He founded the first agricultural experiment station, developed instruction in agriculture and was Dean of the College of Agriculture and Director of the Agricultural Experiment Station, and enjoyed a position of trust and honor among his colleagues. In recognition of his splendid scholarly attainments, the honorary degree of Doctor of Laws was conferred on him by four universities, viz., Mississippi, Columbia, Michigan and California. From 1910 until the time of his death he was Emeritus Professor of Agricultural Chemistry, but in that period never lost his interest and keen enjoyment in scientific work. He died, January 8, 1916, soon after celebrating his eighty-third birthday. He is survived by two daughters, Alice and Marie Louise, Mrs. Hilgard having died several years before her husband. Their only son died at the early age of twenty-one, a tragic event which saddened the latter half of Hilgard’s life. In the history of that dimly defined realm known as Agricultural Science, still a more or less chaotic mass of erudition, and in its forma- tive stages, few names stand out in such bold relief as that of Eugene Woldemar Hilgard. To those whose way of life has been fashioned in the grooves of that field of activity, Hilgard is perhaps as well known and by them as much honored as anyone who had preceded or anyone who has followed him. His name stands there for integrity and high purpose, for its association with the most attractive of the amenities of polite social intercourse, for idealism in scholarship and a fair modicum of realization thereof, for unswerving adherence to the truth as scholars see it, for indomitable courage, persistence, and perseverance, for un- daunted determination to try and try again even in the face of failure, to lead oneself as well as others in the right ways as man sees them, for an insight amounting almost to foresight, for a gentleness, and a dignity, and a charm withal that taker in those other lights bespoke rare balance, poise, and an accentuation of the unusual in man. His name, too, stands 594 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. IIT, No. 4 to his erstwhile fellow workers for the exploratory temperament in its most praiseworthy expression, for the blazer of trails, for the fearless pioneer and delver into the regions beyond the ken of man. And yet that same name speaks to them of simplicity, of a true humanness, of an ability to lead and instruct which have gained its quondam possessor an unequivocal place in the respect and affection of his fellowmen, and the generations which follow. Hilgard was a man of about average height, slender, graceful in car- riage and courtly in manner. Broadly educated beyond most of his contemporaries in science, his excellent classical education and his master- ful knowledge of several languages rendered him a deep thinker and a conyersationalist of singular charm and attractiveness. By birth and breeding a gentleman, he won the hearts of his friends and acquaint- ances by his inimitably delightful manner in social intercourse. But like all forceful men he had his other side. He was possessed of a grim determination, an unyielding tenacity, a bold combativeness which those who engaged in battle with him will never forget. In espousing and championing a cause he gave himself to it in unstinted measure, and fought for the truth as he saw it with an energy, a whole-souled intrepidity and alertness which will always be remembered by his adversaries. It might be said of Hilgard as Ingersoll said of Roger Conkling, ““He knew his friends; his enemies knew him.”’ Thus was Hilgard a leader in his generation. Kind, charming, lovable, he was tireless in spite of a life- time of delicate health, unafraid, unflinching in his advocacy of a prin- ciple, a theory, an idea. That his masterful, tenacious, forceful side at times made him overzealous and occasionally blinded him to some side of an issue is only what one would expect of men, however strong, but essentially Hilgard was possessed of a blending of strong characteristics, rare even among leaders, in which we may truly say he had the qualities of his defects. He was a man through and through, and in his long, event- ful career we glimpse a vision of a life in his home as well as without it which men may well emulate. A devoted husband and father beyond the ken of ordinary men, Hilgard loved and was beloved and was inspiration. As a teacher Hilgard was kindly, patient, resourceful, and devoted. He loved the seeker after knowledge and spared no pains to give to the earnest student of his own ample store of erudition in unrestrained measure and in impressive manner. The educators of today may well study Hilgard, the teacher, to learn the value of a broad education to the scholar and the builder of scholars. He enjoyed teaching, yes, he reveled in it. He was to the manner born. As a scholar Hilgard belonged to and perhaps was the last prominent representative of the “naturalist’’ group of the last generation. And yet 1920] LIPMAN: OBITUARY ON EUGENE WOLDEMAR HILGARD 595 he was beyond that generation since he specialized considerably. Trained successively as a chemist, geologist, and botanist, he gave the last forty years of his eventful life to the study of arid soils and their relations to plant growth. Beginning with his studies of the gases of the candle flame, upon which he wrote his doctor’s dissertation, which was one of the bases of Bunsen’s subsequent invention of the Bunsen burner, he evinced thus as early as 1853 that curiosity about Nature’s unknown which grew in intensity with the years and rendered him an ever absorbed investigator and supporter of new research projects. I may digress to say that it was my good fortune to see the little pamphlet which was Hilgard’s sole surviving copy of his doctor’s thesis more than a half century after its publication and to discuss it with its author, who, as was customary and habitual with him on such occasions, gave me an hour of most charming reminiscences of Bunsen and other professors in Ger- many with whom he had worked. I recall particularly in that connec- tion the characteristically delightful glee with which Hilgard told me of his final examination for the doctorate, in which the first question that Bunsen asked him was, “Herr Hilgard, was ist denn Methyl- alkohol?” Unfortunately lack of space forbids my recounting here the long series of researches remarkable for their day which won Hilgard the distin- guished position which history now accords him in the annals of science. Suffice it to say that whatever he touched he illuminated, and this was especially true in the field of soils, in which he was a pioneer, and to an understanding of which he was until recent years the most gifted contributor. His two hundred and fifty scientific papers, his two quarto volumes on the cotton investigations, published by the Census of 1880, and finally his celebrated book “Soils: Their Formation, Properties and Relations to Climate and Plant Growth in the Humid and Arid Regions’”’, published in 1906, attest most eloquently his energy, clear-sightedness and ingenuity, as well as his charming literary style. The latter was made possible by his broad education and an inherent power of verbal expression, a combination possessed by few scientists. Whether or not the results of his investigations, particularly in his chosen field, remain a living and true picture of the subjects they sought to illuminate, is beside the point. The work of very few men in the history of all science, if indeed of any, stands authoritative, at least finally so today. It is no detraction, therefore, to say that most of Hilgard’s results of experi- ments on physical and chemical soil problems, including his well-known experiments on alkali in soils and its relations with plant growth, stand now in question in the light of the stupendously rapid rise of more accurate investigations in chemistry and in plant physiology. The best established and oldest doctrines in science are even now giving way 596 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS |Vol. III, No. 4 before the remarkable discoveries of recent years, and even months, in the domains of physics and of physical chemistry. But let it not be supposed that this fact dims one whit the lasting value of the service rendered to science by those masterful scholars whose results formed the woof and warp of “modern” science for so long. Even so it is with Hilgard’s work in his chosen field. His results were indispensable steps in the upward and progressive evolution of that branch of activity which is speedily establishing its right to be considered a science. Yea, they were not merely indispensable steps, they were bold, long, and numerous steps. Thus in my humble way and in the most general terms I have tried to delineate the characteristics of Hilgard, the man, the teacher, the scholar, the leader. A thoroughly human, intellectual being, he was a grace to science and an ornament to society. He lived wisely, observed keenly, worked intensely, fought his battles bravely, loved and was beloved warmly, and died nobly. The manner in which he did all these things in the face of seemingly insurmountable obstacles through a very long, useful, and honored life must always be an inspiration to men. Requiescat in pace. Cuas. B. Lipman. GEORGE EDWARD PATRICK. On March 22, 1916, after an illness that kept him from his office for only a few days, died Professor George Edward Patrick, M. Sc., Chief of the Dairy Laboratory of the Bureau of Chemistry, at the age of sixty- four years. Son of Delano and Mary (Maynard) Patrick, he was born October 22, 1851, at the Hopedale Community, founded by the Reverend Adin Ballou near Milford, Worcester County, Massachusetts. After preparation in common and preparatory schools, the subject of this sketch entered Cornell University a few years after the opening of this institution, and was graduated with his bachelor’s degree in 1873. The following college year he served there as assistant in organic and agricultural chemistry under Professor George Chapman Caldwell, one of the original members of the Association of Official Agricultural Chemists. At the same time, Mr. Patrick carried on post-graduate studies, and, upon the completion of the college year, received his degree of Master of Science from that university. Immediately thereafter, he became Assistant Professor of Chemistry at the University of Kansas, where he remained for a number of years, during which he succeeded to the full professorship in chemistry. There also, on June 19, 1879, he married Miss Hattie E. Lewis of that city. In 1883, having become 1920] FREAR: OBITUARY ON GEORGE EDWARD PATRICK 597 interested in metallurgy, he resigned his chair to accept the position of superintendent and manager of a mining and smelting company. Later he served as chemist to the Bradley Fertilizer Company of Boston. But in 1888, upon the organization of the Iowa Agricultural Experiment Station, Professor Patrick was asked to become its chemist, and in 1890 assumed the duties also of the Professor of Agricultural Chemistry in the college. The public demands upon the technical members of the experiment station staffs in the early days of those institutions prevented much specialization in the work of any of these officers. The bulletins of the Iowa station present many studies, individual or cooperative, by Professor Patrick upon the composition and nutritive values of various forage plants and other cattle feeds, upon the chemistry of the apple and the apple tree, and upon the sugar-producing qualities of sorghum and the sugar beet as grown in Iowa. The dominance of the dairy interests of the region soon led him to devote the major part of his laboratory effort to the chemical problems relating to that industry. He left the Iowa college and station in 1895. The following year Pro- fessor Patrick was appointed an assistant chemist under Dr. H. W. Wiley in the Department of Agriculture. Here he remained, devoting his attention exclusively to the examination of dairy products, and, upon the formation of the Dairy Laboratory of the Bureau of Chemistry, was made its Chief. Professor Patrick was a master analyst of dairy products. We are indebted to him for a number of contributions to analytical methods and appliances. The use of the copper distillation flask in the deter- mination of nitrogen by the Kjeldahl method was introduced by him. We owe to him the modification of the lactocrite method for the deter- mination of fat in milk, a modification later known as the “‘brine test”’; devices for the convenient sampling of milk and the measurement of acid for the Kjeldahl and Babcock methods; also, the perfecting of the Roese-Gottlieb method for fat determination in dairy products. This association is deeply indebted to him for numerous comparative studies upon the determination of various components and ingredients of milk products. He was quiet in taste, did not seek the lime light, did not rush over- easily into print, and in the association meetings the frequency of his participation in debate fell far below the measure of his learning and experience. His judgments were formed carefully, expressed most positively, held tenaciously; yet his mind remained open to new truths. His generosity of character is beautifully illustrated by the hearty promptness with which he accepted and publicly commended the Bab- cock milk test upon its appearence just after, by several years of effort, the Patrick brine test had been perfected and brought into use; also by 598 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 his treatment of his assistants, of whom he nevertheless demanded painstaking accuracy. This generosity was balanced by a high sense of justice and courage. Domestic in tastes, spending his evenings at home, he was reluctant to assume duties which, for a time, separated him from his beloved wife. It was a source of deep grief to him that a long period of duty upon an important dairy investigation in the West had separated him from her for months prior to her death, which occurred in Denver during a brief reunion. This event shadowed all his remaining days. WILLIAM FREAR. {gg EE TT EE Se VISE THOMAS CUTHBERT TRESCOT. I well remember the day in the summer of 1884 when T. C. Trescot first appeared in the old basement laboratory of the Chemical Division in the old red brick building of the Department of Agriculture. Born in Charleston, S. C., October 1, 1857, of a family possessing large landed interests but later impoverished in capital by the Civil War, he found it necessary to make his own way. His father, William Henry Trescot, lawyer and Assistant Secretary of State in the Confederate cabinet, was a man of high qualifications and great distinction. He attracted the attention of James G. Blaine during his term as Secretary of State, who sent him on a number of diplomatic missions requiring great skill and tact. He was accompanied on these missions by his son, Thomas Cuth- bert. Trescot thus gained a wide knowledge of the world and its ways. In the old days there was no Civil Service examination, and Trescot came to the old laboratory in the basement of the brick building where the Secretary of Agriculture still has his office, with a note from the Commissioner of Agriculture to Dr. Wiley to “put this boy to work’’. It was soon discovered that he had no knowledge of chemistry, but he had a great talent for work, and especially was he careful of all details of the work which were entrusted to him. Naturally, the work given him was of routine character, but he had a genius for routine. His knowledge of chemistry was never very much extended, but his knowl- edge of the best methods of determining nitrogen was perhaps not equalled by that of any other chemist in the world. He gave his whole service to that branch of chemical science. He began with the old soda- lime method, in which he became a master. This was followed by the old Ruffle method, in which he also became skilled. Then the moist composition process of Kjeldahl came into vogue. He knew that through all its variations of every description from beginning to end. He acquired a delicacy of technique, which was rarely secured by any other operator. By personality and craftsmanship he made his place, and when, with the 1920) FREAR: OBITUARY ON THOMAS CUTHBERT TRESCOT 599 growth of the Bureau of Chemistry, a nitrogen laboratory was organized, he was made its chief. The members of this association who read with care its proceedings from the earliest days will appreciate how much we owe to Mr. Trescot’s constant and careful aid in the cooperative study of the various nitrogen methods. What a unique experience— twenty-seven years of stirring life and seeing the world, and then thirty- two years in a little room making the same determination over and over again, and always with the steady aim of the craftsman at quality and quantity. But the cheeriness and the playful cynicism remained peren- nial, and always the self-respect without remotest tincture of self- conceit. In fact, I have always suspected that for Trescot the day up to 4.30 p. m. was the period of chores to be neatly and unfailingly performed, while the remaining hours were for real living, whether the life of the lover of sports and young clubman, or, in later days, when the right girl had come into his home, the life of the lover of the home hearth and of the proud father. Mr. Trescot was married June 19, 1905, to Grace Matthews. Their daughter, Elizabeth Cuthbert, was born January 29, 1907. Their second daughter, Mildred Carlisle, was born May 4, 1911. Their son, Thomas Cuthbert, was born January 5, 1915. Trescot died April 14, 1916 from heart trouble. Of his own ambitions he breathed no word, over his disappointments, was never heard to sigh; but let a comrade experience either joy or grief, and the shell of cynicism opened to let a warm heart peep almost shyly forth, and the voice that, but an instant before, had been joyously railing grew husky with sympathy. The esteem he won from his associates was manifested by a rare occurrence in departmental circles, a banquet celebrating the completion of his twenty-fifth year with the laboratory, an occasion shared by many professional friends widely scattered over the continent. I can not more fittingly close than by adopting the language of one of the toasts of that evening and dedicating this sketch to the memory of “‘a rare good friend, a man generous to a degree, ready always to help the under dog; in short, to a kindly, courteous South Carolina gentleman, whom it is good to have known”. WILirAM FREAR. Attention was called to the fact that the members of the Board of Editors were appointed originally for one, two, three, and four years, with the understanding that succeeding members were to be appointed for four years, and that a vacancy existed which had not been filled. A motion was made, seconded and duly adopted that the appointment of 600 ASSOCIATION OF OFFICIAL AGRICULTURAL CHEMISTS [Vol. III, No. 4 members of the Board of Editors be referred to the Executive Committee with power to act. Mr. H. C. Lythgoe moved that a rising vote of thanks be given to the president for the admirable manner in which he performed his duties. The motion was unanimously carried. It was moved, seconded and adopted that the next meeting be held in Washington, D. C., and, if possible, at the New Willard. The convention adjourned. INDEX TO VOLUME III PROCEEDINGS OF THE FIRST AFTERNOON AND LAST TWO DAYS OF THE THIRTY-SECOND ANNUAL CONVENTION, 1915, AND OF THE THIRTY-THIRD ANNUAL CONVENTION, 1916. PAGE Acidity, titration in colored solutions, paper by Hartmann..................... 410 Acid soils, lime requirements, paper by Conner..................222200000002 139 Address by Vrooman, Acting Secretary of Agriculture..........................0-. 418 Gishlonoranyperesidenta by, WEY. <. socio ecole erties sien ce einen 420 anlteration £000 sreport by HOLtyet.,. =..5.550 scsiecneteee oe nce cinclion em ce ee 398 Alcotiok:tables, paper /by, St: John, reference. ...... 2... 022 220+ccescnssecnceccies 197 Alkaloids PEGOMIMEN GE LOMB PONBEMIN EL. sss 2275/3 6 eve: s. cxersier ernie eine tie ite oimiee 193, 380 ECDOLUO NRE UCR So. oi< eese) 21525) 0, oc s)aseis cisissouge cee ceo eerie renee 188, 379 Alsberg finanGialestALeMmentiens OUTTAL: ... o5.5.2/5 sore fase Croce oiaia dolanticee as eee 576 report of Secretary-Treasurer forsyearjending, November. 17) 191555 2 o.. cee= eine ecrcise eee leneie = 54 forsyear ending: November 22) 1916 )21. l.ys tls. -loieiebeter ere eer 197 602 INDEX TO VOLUME III PAGE Barnard, report of Committee C on recommendations of referees and revision of MOEtHOUS SS Se icra io osaes & wate tale varch pee aise Seiede Nau Role vas vase here tetole ste teasers 72, 528 Bartlett, report, teaiand 'cotlee ge eicrieeirc elem chic elie tit cies ic clare sicisieierele ie nena 21 Basic slag availability of phosphoric acid, report by committee (Williams) ......... 104, 585 phosphoric acid content recommendations by jyKeser. cece ier oe eeienieecie cet is eiarielsre taints 288 Teport by ykes iby. fi we theevere otercyetchernye lesa tena eerste ey alsatoge ee ee 193 estimation of strychnin in presence of quinin, reference. ...............-.-- 193 By-laws and constitution, amendments, report of committee (Ross)............- 586 Cacao products changes ini methods).1.-...<:..< 5 = ae geueis erya ard Soe ieee + oie eine eee eee aya! See also Cocoa and cocoa products. Canned vegetables. See Vegetables, canned. Cereal products Chanwes I MeCCDOMS eee cis. <5cia.w) sos lieievecskerete: elepe PACES a Lelekeheyan~) Sa ciey este een 558 method for measuring acidity; application to sulphured and unsulphured oats, paper by: Bincknersreference «p96, cm deaniele mpl) = shes seis ee eae eee recommendations By Commntbee ge. coat eis Se srestesatesa seus oa teleteretetaeltiela oie: iy niky stp itet Race 87, 532 byrlie! Gleres sane seis ce aici cle onc hie ete OMOR ene «er oe aint 451 report by: Le Glere.G ice pissy ot ccs Slice iat ar Seay ENP lS lene 446 Cheese and milk, separation of nitrogenous substances recommendation by Bosworth. ccc piccus ssc ieiaucte onebpeseistaskale Reeicpushe Pie eet ene 53 hy: Gommittee vB ays eves. cie:5 oittetcre ehote asso onal sso) otel slnindassy=/= ete eRe hee 71 by sPalmers sac cece en's oltye sya Waele te oo Tee tere letereke etre, See erel tee tare 274 report by Bosworth s jas. seivicissconcwe mie otestreloieie ous geese = ett nee 53 by! Palmer’: ..0-055. 528 cl eds scoane © aeistereteten slate O Rl als eer 273 Chemical reagents, testing recommendations hy) Comittee 1B oii. iis isis cere ate sisters evel s‘ohotote sus] eiotnayenathiata Niele ieT« [saa 71, 528 EN@l 2), 21:1 Ha HOBIE OAS eOCEn ss ceianods CTdeo oo aon nAsGosse: 226, 277 TEPOLt Wy: UWL ieee ie vast cic eantancoye ehtbe nebed tn) pales gobs tas ae i be te oe 225, 276 Chesnut paper, determination of pepsin in liquids, reference............-...-.+0500> 197 FY: 0) i OVO eee OMT SO OD JOO Sea iononoca Sabo ieys 387 Cider, changes during fermentation, storage, and conversion into vinegar, paper by Hartmann and Tolman, reference.................cseesceseeecserrceess INDEX TO VOLUME III 603 PAGE Citrates, use in determination of phosphoric acid, report by Jones.............. 97 Cocoa and cocoa products Changessimemennouswenpier. sit ot oes «sci s as Mate nee te Re lace 571 recommendations ley LUO? 515 pCO Se DED Oe OO De HORSES oe re ne en ano eee 497 lon (CAyFTITNFEED Chay cles ene ieee eR eaten cok (4 UR eas NEG oe oe 535 ME POCHIVEES IOMMUCL LEE ey. oneye eso ciciziomil ayes PELE Cet ae ne ee 486 Coffee and tea ZUG pHOULOMIMeE MOUs aft: lid Lsi.d,.ler SER derega eke Seo 571 recommendations IVE ArCL TMP ye) oso « «213.02 tac Ao ee ee ee ee by Committee C 5 ISN PUAOMUISHOP EE eerie 25 2 > , ciscateie teas «doce ecb eye aos hiss ao 555 LECOMMMENC AMOS TU Vs OOMMILLEE Ci... ao cis-c occ eee nls = is ae Cee 72 LE pOL tag Viale WSONMERIEETS «.y8ic(scs).) eet okt, SE eee la steely ii 400 LALGCA UE MUADCLUDVIDOVIC. - oo! < o ne oat TO eo dra ole bene 402 Committee A on recommendations of referees and revision of methods, report (SIRT GH) 5 oc upside toi PO aa ee ea) ea Ks) Committee B on recommendations of referees and revision of methods, report (STALLTE TS) «no ese 9 aE nS ALSO ORISSA at Ll AR aie 08 66, 524 Committee C on recommendations of referees and revision of methods, report UES ALZLAECD) Pe CP ena foe a wide Oe Oe ae 72, 528 Committee of review on analysis of lime sulphur solutions, report (Davidson).... 188 Committee on amendments to constitution and by-laws, report (Ross).......... 586 Committee on auditing BpPOMLeN UAC personnel: os). gcc | + CeO tee ee ene 53, 279 FREIOELT be 2 o.q) Geka PAS TCU aeRO Oe CORE RE Ser Ces Hinl ah teens ere ore Af 7 bs% 577 Committee on availability of phosphoric acid in basic slag, report (Williams)..104, 585 Committee on editing methods of analysis, report (Doolittle)................. 89, 537 Committe on food definitions and standards, report (Frear)................. 198, 580 Committee on nominations AMPaUMEN MENG PErSODNEl) ¢ <2 sce -tsjsie-e/s, 0s OBOE EEO fee cee ee 53, 279 MEPOLMDHICUD) Seer iae case: felt siaeis ses ec cletioere erence re oe Lee 187 ECPOLEM VIL DI OLKIEIL) ope oo aie chee eee eee lie ae ee 589 Committee on resolutions UpPoOMUMentrand Personnel. «2 Aes. \ais = ss cs esete hein = gash dees ee 3, 279 FED OLIMOA TEC SOU) Spore vic <5. 6 a hale eis cn ckeus ohe PET ice etesta Ss ceayn ea Sic ERP ete 187 TE DOLUEUIETCAL) Poet ont te. = sca aaveue hecite cate ota ete cere eno eee ee ese 589 Committee to invite Secretary of Agriculture to address convention, appointment LIMIT ROLE re eel e 0 cic cis noe. Soostehe yin, aos wre irae ees Sr ols EL aaa a te 279 Committees, appointment and personnel................600-0 002 ee cece ee eee 53, 279 Committees, officers, referees, and associate referees, for year ending November on Condiments Chaves yMetlhOds:, -s.5. 2 s.2.ece says ites ee Tere ER ee Cie ee Maes ods > x « 569 Recommendainous by Committee) Gy ce cost oe isey is Fore ee actor 535 Conner, paper, lime requirements of some acid soils....................00.0.-. 139 Constitution and by-laws, amendments, report of committee (Ross)............. 586 Cottonseed meal, inosite-phosphoric acids, paper by Rather, reference........... 254 Crude fiber recommendations ss Jeny CUrTITI Ty POR eee eo te Se REGU Se >. ioene ccm pierre cement chscrs 525 WAR TANGCISS cc sc. oee, = 222 cs cise cainilala wie © Goi saves eae een aKee eeoeme eke 260 MEPOLE DV AB TANCISS (|=. 505) 1s) stauct vols sisieiairieloisl one = Sialate fle ae 548 recommendations by Gommnittee Aco. 2 occ als a eeieieie vinis cee lo ciele eo ene ete 62, 521 Jarrell paper, substitution of sodium sulphate for potassium sulphate in Kjeldahl- Gunning-Arnold method for determination of ammonia in fertilizers... .. . 304 report determinationiof: potash. ; .:.<; 5p sciem so ors Sita ease anaes ele eee 107, 315 Jones, report phosphoric Acid acwjcan rere as s150 «ee See 8 CASED oles 56a eee 279 use of citrates in determination of phosphoric acid........................ 97 Journal financial statement (Alsberg)). a)... cunt ahovs soahs Reis cso. as + yekeiae seis 54, 576 report by Asher gsc oie ceo. «bres in: «. vevecataievovs, cyercis aise peienels (= ove coke teeters Relea eae 55, 578 Keitt and King, paper, new, rapid and accurate method for estimating lime and potash. inisouls) TefEreNCe Soa, «,oseiers Jaye histeravels is aselare oe eer «isla ee 326 Kerr). reportafats' and ost.) ..2o)).. 2 soos GER. Len ere eee csi he ene 432 King and Keitt, paper, new, rapid and accurate method for estimating lime and potash ‘in ‘souls}’reference’. 6h WEA. LI Pe a eee ones ek 326 Kjeldahl-Gunning-Arnold method for determination of ammonia in fertilizers, substitution of sodium sulphate for potassium sulphate, paper by Jarrell. ..... 3 Kjeldahl method for determination of nitrogen report by Phelpsrand Daudty sic succysretelere: fteueee sol ale Cal antegstots ings aro Renee 218, 306 use of potassium permanganate, paper by Frear, Thomas, and Edmiston. ... 220 Klein; report, metalsin food ..,:;, ai... seats bee we eather eyo ean tale ene 512 Kilueter, report; dairy products; .:0.:'ee est sp ecleolaiiene ete otein es iia 254 Kuzirian, paper, separation and gravimetric estimation of potassium............ 321 Leathers: changes: inimethods:. ubsEbaGoooseo 2ostis Hoc 53 ya Committees Bi 35%. ae oerne eh ee seers serene a etter a: 71, 527 Bong all mer os chet spss oerasoversaeieysiercr= c,ch ata aoe eI ae eee forecast 274 report liny LS aS Oe OO ROPE Sa eID ac oot on AoA e bande som rice 53 lohy IPAIIGie ss Spa opeoaes Ae DNS eEOSCOnS Acca dec Coccgor abt ooost SeoceSe 273 Milk, cows’, neutralization precipitate, paper by Palmer....................... 274 Nitrogen determination Mifertilizers; paper by; Lrowbridge.. ~.2 62s. cane 2 eee lee eels 217 investigations of Kjeldahl method, report by Phelpsignud Daud ties eve oes cine ote secrete er 218, 306 FECOMMENGALIONS; Dy; PMCMLEL tele le oe )ayayo rein eile os elavade) =< tenella eteterete te -ii7r= 304 608 INDEX TO VOLUME III PAGE report by Deemer so 5 on)c isin ck ese oe eae canprkegehs cheno Miele we aloie take ee 299 use of potassium permanganate in Kjeldahl method, paper by Frear, Dhomas/ and dmistone sce ck os eee eres, 125s perder eee 220 recommendations by Brackett, anG: EaskiS)

    274 report by Bosworthe ese cae eties ee OOP eRGne ts > +--+ ee Ener 53 Jeng iV tile esate ae HFG OA Genco 0 ji: SE DOMO Pc Sod 28dc05 273 Nominations committee, appointment and personnel. .... 0.0.22... 20.00.00 eee eee ee 53, 279 report, of committee (BLMELL) <6: ose. te see eens 2 <> ee eens 187 report oL.commuttee (McWomnell) 3.5.4. aspen ei oe eee 589 Noyes, paper determination! of. phosphorus SOmUss.) 42.) ie 2 + © clea erste ee 149 study of soil containing residual limestone... ......5......06--seeeee scene 151 Oats, sulphured and unsulphured; method for measuring acidity of cereal products, paper by Birckner) references ¢ 2/.::5. \ tiie. 5.s 6+ cpesiiere wikia > lovee ete ete eet 255 Officers, referees, associate referees, and committees for year ending November AGUS. dsc dic ace da nine PHSeE 2s tA TERE mt DELO te cece nta rane 228 Oils and fats changes:in methodsy.cs..c5c ccs crete os bt CRM RGRIE Sen OA Skies «bea eemee 568 recommendations by (GommitteciGiy .cassa ines oF in ant he Seem C hottie ete ete onan 86, 535 by: Kern.) ccsme 66 56 ctv it 6.4.5 0a 4 bre hm eee CIC NaTNNTs Nee Ve noite) set etna 436 reportiby), Kern. scmnicinte Sie le ciniel« «oi 4 gletel out ieee eile «te ee ie ee 432 Optional method, defined ix... say onc tA sls cinta erttotete picts cle cle Retains i eae 227 Palmer paper, origin of neutralization precipitate of cows’ milk..................-. 274 report, separation of nitrogenous substances in milk and cheese...........-. 273 Pan-American Scientific Congress, representatives named...............-...55 53 Papain: report by Chesnuts << 5. (cise ne eine © taieiets’> cane oivlal h-eiettnGe)= tis ire 387 Patrick, George Edward, obituary by Frear................---.sseseeerversse 596 Patten, A. J., report, inorganic plant constituents.................-+.0e5005 153, 329 Patten; H.,E., report, ‘baking powders. 00's. 2). + +s scam stone sea eet Rieti Paulsreport, dla vOringyextracts:. «oye siete iciete ees = o/s oie a elm = sce Ree re Ienegte 415 Pepsin in liquids, determination, paper by Chesnut, reference .................. 197 Peru balsam. See Balsam Peru. Phelps and Daudt, report, investigations of Kjeldahl method for determination of WACOM EN 315 Ne te PRA SIIS ars, 3 dhis bine Ses oln Ril» an let tet PS ONE io Renee is i eee 218, 306 Phosphoric acid recommendations by Gomamititee vA:.isi.t .aijaitia sinner ele, Sas Dy Abas, SEs hates een 59, 519 by Jones ile 22 seta 2. 2:3taterwiw te ss ahe ne a\iwin abeye Paty eh le ate Na lene =) sts ea ees 286 Dy Waller sich cit2 Jay ues seit salva «a oute asics he) bss Bo sie Wi cs en 95, 96 INDEX TO VOLUME III 609 PAGE report Ly CIDE. Soto oo Gb So Oa IEREREREIEN © Cf eCEROTE Re am See es a vin UE 279 TS LLSES. 7 . 23 aa ae al ee Enid Alt OOF alt 90 use of citrates in determination, report by Jones.......................... 97 Phosphoric acid and potash, preparation of organic material for determination in aliquotsiofesame: solution, paper by West:...........-..--.-sctee eee eee 99 Phosphoric acid, availability in basic slag, report of committee (Williams). ..104, 585 Phosphoric acid in basic slag PECOMMMENU AMON WV) UiVKOS 2, «<2 orre ces 4 ys ose ert occas mn ee 288 mera Liny Lag kee ac Se ee ein ap annie OEIC aad eh OAR Soe a ata 286 Phosphorus inorganic in animal tissue recommendations by Forbes... : 0-0: heen ee eee 273 report Bye Beer lens 75) ctsoays oun reteset lahore ee 204 BU OPUS ret ds, 533 nnd eee eee ee eee 264 msouss determinations paper’ by INOYeSon..00o. sae ee ee eee 149 organic and inorganic, recommendations by Committee B ............... 70, 526 Plant constituents Charipespinume LUOUS PEE ees: se a< « eicie som arene nt aia ree Oe Te 544 inorganic recommendations ID VAGONIINILLEG ASS. sisinss nese ee ne cece oe eee ee 61, 521 i VpEAGtGU papers cies iss cieard Gags ioc ie were Oats teree ne ee 156, 331 Oat. logy LAAN as ee ee Oe ae eS cnn oe NN Cae Soe Road 153, 329 Plants and drugs, medicinal recommendation by, ommittee Bo. 5. <.. 2s sas as use cna seer iter 71; 527 FEeCoMMeNnGALONs DY) VICNOCVET;: |:..,.,0.<(1:s.204 aE eo eo Ae era 386 MEpOKtDYpVICHOGVER-~ ss... \-/eaciaemtint.atek A asiellh Guagete sere eevee aes ahyercre ered epabters 381 Potash availability in commercial wood ashes, paper logy LAGOS Coc Oe ero Se POMOC oor cries sable ome ceue 323 bypotallngs.and Wilson, reference. - saci... « «a> = «= ) hierar 88, 528 bynoeek ers: eases emis err iesitie iets tani ets ee 38, 511, 512 by seeker anid: Wolf cic -rcyetr ets foyetey iat eh tepeien -ledeteh cision -lotel piel yotey 2. 42 MEDOMMIDY; SECKEL laine ih.) miel astey si siefaico) 2 eefeis 2) iin erika ease eal ogeeds a 33, 504 previdentisiaddress| by prackett ... 5-5-1). -\: 4)-\-. sleeiplete ie soe Fa oiractd ie eferete 238 TE enmeriadl iran res Bit G Hie Beep Sirs ee eriooe OSE enn Rae ho sed dae caro cis mace 227 610 INDEX TO VOLUME III PAGE Quinin, estimation of strychnin in presence, paper by Buc, reference............ 193 Rather, paper, inosite-phosphoric acids of cottonseed meal, reference............ 254 Reagents, chemical, testing recommendations by Gommibtee Big) hc pies auch ia opel PARR re) for cas vp ee 71, 528 Dy, EWI ge i255 J sate evsceys! Natero srl siesta eRe eee eit te kepeitees 226, 277 report bys Biwings 7ko Se ako hates ee ee ne eee eee 225, 276 Recommendations of referees and revision of methods, report of Gommirttes cAb(Skinner) e205 sets hot re eee ote oe oe ‘ofiGommittee sb (Stallinvs) 77. 4ser acre onto ee ee eee eee 66, 524 ‘of(Committee!G\ (Barnard) )is-1s genet alee eo aca ee 72, 528 Refercess distribution TevIsed sas wise soos oe ae ee Fie eee 579 Referees, officers, associate referees, and committees, for year ending November 1917. 228 Resolutions committee, appointment and personnel..................cc0ceccceeevee 53, 279 report oficommyttee\(Davidson) ss. 2. oie eey- Gk le oe «Guin. eee 84, 532 Teport Dy areas. 122 ssi ec.) Ree one a CRs «sci na ea ee 409 titration of acidity in colored solutions, paper by Hartmann............... 410 c pone sf Ss: r ¢ “= n <= oF e ‘Ss Association of Official 583 Analytical Chemists AT Journal VES J Bio! > rea & N le: lica] Serials PLEASE DO NOT REMOVE CARDS OR SLIPS FROM THIS POCKET UNIVERSITY OF TORONTO LIBRARY Si ORAGE