16 DAVIS AND LIR'G: ACTION OF MALT DIASTASE ON By BERNARD F. DAVIS, B.Sc., and ARTHUR R. LING, Introductory. IT is well known that when starch paste is hydrolysed by a n aquems solution of malt diastase, the speed of the reaction as indicated by the diminution of the specific rotatory power, and the increase of the cupric reducing power of the products in a given time becomes less the more the temperature of hydrolysis exceeds a certain point, which is termed the optimum temperature of diastatic action. Brown and Heron (Trans., 1879, 35, 596) give numerous determina- tions of the specific rotatory power of the products of the hydrolysis of potato starch in the form of paste by malt extract at different temperatures after the lapse of definite intervals of time, and from their results i t may be deduced t h a t the optimum temperature of diastatic action lies between 50" and 60'.Kjeldahl, who, in the same year (Cornpt. rend. hbor. de Carlsberg, Copenhugen, 2, log), confirmed this conclusion by determinations of the cupric reducing powers of the products in question, found the optimum temperature to be 55". The velocity of the reaction between diastase and starch paste, if measured by the constants just referred to, becomes less, pcwi pussu, at temperatures above 55" up to the point at which the activity of the enzyme is permanently destroyed. Moreover, if solutions of diastase are heated above 55" before mixing with starch paste, the same apparent slackening of the reaction or weakening of the enzyme is observed. The effect of heat on diastase is usually spoken of a s restriction, a term we have adopted, it being understood t h a t in all the experiments recorded in this paper the diastase was invariably previously heated in aqueous solution.In a paper read before the Midland Counties Institute of Rrew- ing ( J . Fed. Inst. Brewing, 1902, 8, 475), we gave an account of a large number of experiments on the action of diastase (previouslyPOTATO STARCH PASTE. 17 heated in aqueous solution and in the dry state, at various tempera- tures for definite intervals of time) on potato starch paste. In the case of previously heated diastase solutions, we found t h a t when the temperature emplojed was not above the optimum point, the reaction with starch paste new the same temperature (not above it), provided a sufficient mass of the enzyme were employed, proceeded rapidly for the first 7 5 minutes, and then advanced steadily until, at the end of about 40 hours, the specific rotatory and cupric reducing powers of the product were ,zpproxima.tely those of maltose.When, however, the diastase solution was previously heated above the optimum temperature, the reaction was not only slower, as has been obrerved by others, but the products were apparently different. Thus we found t h a t by the action on starch paste of diastase solutions, restricted in this way at 70' (the hydrolysis being allowed t o proceed at 55'), d-glucose could invariably be detected afher the reaction had continued for several hours. On the other hand, the products of the transformation of starch made with diastase solutions which had been previously heated f o r 30-60 minutes at temperatures not exceeding 55" were found by US not to contain d-glucose, even when the reaction was taken t o its final stage.Nor could we detect d-glucose in the starch derivatives made with diastase, which had been previously heated in the dry state a t temperatures as high as 1 2 5 O . The only change in the enzyme observed in such cases was a weakening of its action. It therefore appewed t h a t the production of d-glucose was connected with the initial heating OF the diastase solutions a t temperatures above 55", and since the publication of. the earlier paper (Zoc. cit.) the work has been continued principally with the object of obtaining further information on this point, The diastase used in the experiments was prepared by C .J. Lintner's method (J. pr. Chem., 1886, ii, 34, 378), and, except where otherwise stated, from Odessa malt dried a t a temperature not exceed- ing 33". W e are indebted t o llr. R. E. Free, of Messrs. Free, Rodwell and Co., Mistley, for placing at our dispxal several samples of this low- dried malt. The starch used was the purest potato farina. The temperature of hydrolysis never exceeded 55', but in a, few cases it was a degree or two lower than this. The methods of analysis adopted are explained i n our earlier paper (Zoc. cit.), with the exception of t h a t used t o estimate d-glucose, which consisted in weighing t h e phenylglucosazone produced under standard conditions, and calctdating from the weight the percentage of the hexose on the total solid matter in the solution, The values obtained are denoted by the symbol G,,,.The symbol RaSg3 represents the VOL. LXXXV. C18 DAVIS AND LING: ACTION OF MALT DIASTASE ON cupric reducing power, expressed as percentage of maltose on the total solid matter, calculated by the solution factor 3.93 from the specific gravity of the solution. EXPERIMENTAL . I n order to give some idea of the power of the diastase preparations employed, we may quote the following experiment from our earlier paper, in which unrestricted diastase from t h e same Odessa malt (0.08 gram) was allowed t o act, a t a temperature of 55”, on starch (10 grams) made into paste of about 3 per cent. concentration. Time. [Qln 3’93’ *R,.93’ 114 hours 150.0” 78.8 184 9 ) 143-6 90.6 42 7 7 138.5 99.7 66 9 9 138.0 96.9 138 Y , 137.6 99.5 An experiment since carried out with diastase from the same source (0.5 gram) and starch (10 grams) made into a paste of about 2 per cent. concentration gave the following results, the hydrolysis being conducted at 47” :- Time. [.ID 3.93- R,. 93. 89 9 9 134.6 97.5 67 hours 135.5” 99.6 This solution, which gave well-defined ma1 tosazone but no glucos- azone on treatment with phenylhydrazine acetate, also yielded malt- ose in the crystalline form, These experiments shorn t h a t the reducing power attains a maxi- mum and then undergoes diminution, a phenomenon which we have previously referred t o as “reversion,” but for which we have not at present found a satisfactory explanation.It is invariably exhibited in the case of conversions with both unrestricted and restricted dia- stase, and it may occur at any point of the reaction if the latter has stopped, because an insufficient amount of diastase is present t o carry the hydrolysis to its final stage. I n addition to the cases already mentioned (Zoc. cit.), we have now obtained further evidence showing that not only does the reducing power diminish when the temperature of hydrolysis is maintained beyond a certain period, but that the rotatory poNer increases. The following experiments were all made with diastase restricted by heating in aqueous solution at various temperatures.POTATO STARCH PASTE. 19 Se~ies I. A . Three separate amounts of diastase, (a) 1 gram, ( b ) 0.5 gram, (c) 0.2 gram, were restricted a t 7i*5-7S0 for an hour.They were then added to three 10 gram quantities of starch, each made into paste with 375 C.C. of water, and placed in a thermostat at 52-55'. IDS 93- R3-93. A- - 43 h0~1-s l g i * i o 190.9~ 187.20 6.7 2.6 1.7 91 I , 184.5 191.4 189.7 15-3 4.4 5.8 187 19 181.7 191.1 189.4 18.7 4.5 5.4 Time. (a). ((5). ( c ) . (a). @). (c). - - 20.8 - 259 ,, 183.2 -- (a), on fractionation, gave a dextrin insoluble in 46 per cent. alcohcl having the constants [a]D3.93 = 184-0°, R,.,, = 2.36. A fraction soluble in 80 per cent. alcohoi having a reducing power of R,.,, = 57.3 was separated ; but as this was contaminated with foreign matter from the diastase, and its amount was too small to attempt purification, i t was not further examined ; (6) and ( c ) were fractionated and found t o contain no constituent capable of giving a crystalline phenylosaxone. B.An attempt was now made to carry this reaction further by employing a lower temperature of rest'rictioo. To this end, 0.5 gram of diastase was restricted for an hour a t 75', the solution being then added to 10 grams of starch made into paste with 375 C.C. of m t e r . The temperiiture of hydrolJsis was 52-55', Time. [.I 3-91. J33.93. 65 hours 171.8' 49.5 113 9 ) 172.9 44.7 161 9 9 173.9 47.0 260 ? ? 158.6 62.5 The reaction mould have undoubtedly proceeded further if more diastase had been employed. The final solution, which gave no insoluble osaxone, yielded a soluble os;tzone having the form of " iso- maltosazone" and melting below 150'.This result showed that d-glucose, if formed a t all, had again disappeared (see pp. 25-27). C. Diastase (2.25 grams) was restricted for 15 minutes a t 74", and subsequently for an hour at 75-75-5'. The solution was then added to a paste prepared from 30 grams of starch, and placed in n thermo- s t a t at 5 5 O , the total volume of the liquid being 700 C.C. Time. [.ID 3'93' R9.93' 16 hours 191.0" 10.2 88 7 , 190.0 1 l . T 187 ,) 186.i 16.4 c 220 DAVIS AND LING: ACTION OF MALT DIASTASE ON At tbe end of 187 hours, the solution still gave a blue coloration with iodine, and after adding 0 02 gram of unrestricted diastase it was again placed in the thermostat at 55". Time. [a], 3'93' R,. !)3. 65 hours 1 3 8 . 9 O 95.6 89 9 , 139.7 93.8 After 89 hours, the solution yielded a small amourit of insoluble osazone, but the main portion of the product obtained by heating with phenylhydrazine was soluble in hot water, and had the form of '' iso- ma1 tosazone." Subsequent experiments showed that the period of heating a t 75" was too long, and that the unrestricted diastase subsequently added was not in sufficient quantity t o complete the reaction.Any d-glucose which might have been formed would probably have dis- appeared again. Series 11. The diastase (1.432 grams) was restricted a t 75" for an hour, the solution filtered and added to 10 grams of starch made into a paste with water, the total volume being 400 C.C. After remaining in the thermostat a t 55' for 115 hours, more of the same restrlcted diastase (1.305 grams) was added, and a further addition of 1.572 grams was made a t the end of 238 hours.The quantity of diastase used, which in the final solution corresponds with about 500 grams OF malt, renders the constants (especially the specific rotatory powers) untrustworthy, although blank experiments and corrections were made for each addition of diastase. No insoluble osazone was obtained from the final solution and only a small yield of soluble osazone. When the portion of this solution soluble in 90 per cent. alcohol was evaporated, even in alcoholic solu- tion, it rapidly became brown. This phenomenon, which has been frequently observed by other workers in the case of the products of the hydrolysis of starch by restricted diastase, will be further dis-POTATO STARCH PASTE.21 cussed in a subsequent paper. The alcoholic solution of the final products of the conversion just described gave no evidence of the presence of maltose, and did not yield crystals even after a consider- able lapse of time. Series 111. Attempts were now made t o follow the reaction with lower tern- peratures of restrict ion. A . Diastase (0-2 gram) was restricted for a n hour a t 6 3 O , and the solution added to the paste prepared from 10 grams of starch. The solution, which contained rather more than 2 grams in 100 c.c., was placed in the thermostat a t 52". Time. 3' 9:P 43 hours. 160.3O 91 9 , 148-5 187 ) 3 145.5 The final solution yielded a small amount of glucosazone when heated with phenylhydrazine acetate. As 0.2 gram of this diastase, if unrestricted, is more than sufficient t o convert 10 grams of starch into maltose i n 50 hours, it will be seen that apart from the produc- tion of glucose the heating at 63" has produced a great alteration.Subsequent experiments showed that by employing a larger amount of the restricted diastase it is possible to carry the reaction still further than the point attained in t,his experiment at the end of 187 hours. B. Diastase (0.2 gram) was restricted for la hours a t 60°, and added to 10 grams of starch made into paste. The solution, which contained rather more than 2 grams of starch in 100 c.c., was placed in the thermostat at 55'. Time. [a], 3-93. R3.93' 65 hours 144.5O 88.7 113 ?, 145.5 87.9 The final solution was fractionated wihh alcohol, the portion soluble in 90 per cent.alcohol yielding, on treatment with phenylhydrazine acetate, a n osazone (m. p. 150°), entirely soluble in boiling water and cousisting of a mixture of the flat plates characteristic of maltosazone with the aggregates of '< isornaltosazone." If the evidence thus far adduced be taken in conjunction with the results published in our earlier paper (Zoc. cit.), i t will be seen that when a solution of diastase is heated from 60' to 7 8 O , the enzyme, besides being weakened, is modified, and, in virtue of this modification,22 DAVIS AND LING: ACTlON OF MALT DlASTASE ON different products result from its action on starch paste. The nature of these products remains to be determined, but we have established the fact t h a t d-glucose is formed when the heating of the diastase solution is carried out between 6 3 O and 70".Later experiments show t h a t d-glucose is produced when the heating takes place at higher temperatures (for example, 78O, p. 29) if this is only maintained for a short time. Having thus established the fact that this modification in the action of the enzyme is a resnlt of its restriction in aqueous solution, it became interesting t o ascertain whether this was due to a permanent altera- tion of the diastase molecule. The question also arose as to whether the d-glucose obt,ained when restricted diastase acts on starch owes i t s origin t o the action of the modified enzyme on maltose, assuming this to be previously formed. As a result of several experiments, we stated ( d i d .) t h a t d-glucose is not prcduced by the action of restricted diastase on maltose, but the point is so important that it was again investigated. Xei-ies IV. Diastase (6 grams) was heLrted at 68" for 18 hours with 100 C.C. of water, the solution filtered and made up to 250 C.C. (u) Forty C.C. of the solubion were added t o 15 grams of starch made into paste with 300 C.C. of wat,er. ( B ) One hundred and fifty C.C. of the solution were precipitated with alcohol, the diastase collected, and 0.3 gram of the recovered sub- stance dissolved in water at 5 2 O , the solution being added to 15 grams of starch made into paste with 300 C.C. of water. A blank experi- ment with some of this reprecipitated diastase was carried oKt at the same time. (c> Twenty C.C.of the original restricted diastase solution were added to 170 C.C. of a 4 per cent. solution ( c ~ . ~ ~ = 4) of maltose. The remainder of the original restricted diastase solution was used for a blank experiment. (4. . Time. [.Ill 3.93' R3.93. 19 hours 153.7' 71.4 43 , Y 151.8 79.1 67 ? 9 150.0 79.4 139 1Y 147.6 82.0 (b). Time. C a l D 3'93' R3*93' 17 hours 164.2" 59.3 154.1 72.2 113 Y ) 153-0 72.3 41 9 ) d-Glucose was produced in considerable amount i n both these experi- ments. It is therefore proved t h a t the alteration of the diastase by heating with water (in this case at 68') is a permanent one, as i tPOTATO STARCH PASTE. 23 retains the property of producing a certain amount of d-glucose from starch after being precipitated by alcohol and redissolved in water.It is conceivable that d-glucose may also result from the action of unrestricted diastase on starch paste, but that i t is condensed t o a polysaccharide by a secondary action of the enzyme, Experiments to test this point are now in progress.* Expeyirnent c.-Action of restricted diastase on maltose (see p. 22). R,. 93' Time. [ I D 3.93' / -, 19 hours 154 2" 99-96 100.12 139 9 9 134.0 99-12 99.18 There is no indication from these constants that d-glucose is formed by the action of restricted diastase on maltose, and this conclusion is confirmed by the fact that no insoluble osazone mas obtainable from the solution. The constants (which, as in all other cases, were fully corrected for the effect of the diastase remaining in the boiled solu- tion) differ from those of pure maltose to an extent beyond the limit of experimental error.To what this is due me cannot a t present say. The following experimeut illustrates the action of unrestricted diastase on maltose, giving the constants [alD 3.y3 = 137.8", R3.93 = 100. About 3 grams were dissolved in water, a solution of diastase (0.1 gram) added, and the whole made up to 100 C.C. This solution was placed in a thermostat f o r 48 hours at 55", and, on examination, gave the following values : [ a ] D 3 93 = 131-5', R,.,, = 99.5. The solution yielded no ghzcosazone, but the soluble osazone obtained crystallised in stellate groups of needles and melted a t 169-170O. The maltose, after the above-described treatment, was completely fermentable by yeast. It now became of interest to determine the amount of d-glucose formed by the action on starch paste of diastase, restricted for differ- ent lengths of time a t various temperatures.For this purpose, the sugars in a known volume of the solution mere converted into their phenylosazones under standard conditions, and the insoluble glucosazone collected and weighed. After about 50 determinations, the total mixed products were examined and identified as d-glucosazone. This result, taken iu conjunction with the rotatory and reducing powers possessed by certain fractionated products which gave the insoluble osazone in large amount, proves conclusively that the osazone was derived from d-glucose, and that this sugar is one of the products of the action of restricted diastase on starch paste.* Cornpare Ling (British Assoeiatioib h'eport, 1903, and J. bid. I?&. Brewing; 1903, 9, 450).24 DAVIS AND LING: ACTION OF MALT DIAS'L'ASE ON The osazone, as weighed, melted as a rule at 198' ; sometimes, how- ever, the melting point was higher. After being twice recrystallised from alcohol, it melted constantly at 204'. When i t was dissolved in glacial acetic acid, the solution had a Iaworotation (compare E. Pischer, Bey., 1890, 23, 2119). Maquenne (Compt. rend., 1891, 112, 799) and C. J. Lintner (Zeit. ges. Brauw., 1895, 18, 153) have both devised methods for estimating various sugars a s osazones. After numerous trials, we found that, for our purpose, exceedingly concordant results could be obtained in t h e estimation of d-glucose as osazone by working under the following conditions.Twenty C.C. of the solution, containing 2-3 grams of starch pro- ducts per 100 c.c., are mixed with 1 C.C. of yhenylhydrazine and 1.5 C.C. of 50 per cent. acetic acid in a boiling tube, and heated for a n hour in a bath of boiling water. At the end of this time, the liquid, which has evaporated t o a small bulk, is carefully poured on t o a tared Gooch crucible, and, after the mother liquor is removed, the crys- talline glucosazone is transferred t o the crucible and washed with a small quantity of boiling water (about 20-30 c.c.), so t h a t the total filtrate does not exceed 50 C.C. It mas found t h a t the maximum amount of pure glucosazone could be obtained by one hour's heating. If the heating i s continued for a longer time, decom- position of some of the soluble osazones ensues, and the products of this decomposition contaminate the glucosazone.Under these condi- tions, 0.1 gram of glucose mixed with various proportions of maltose gives 0.0505 gram of glucosazone. I n filtering, care must be taken to allom the mother liquor t o pass through the Gooch crucible before adding the "washing water, otherwise the soluble osazone is precipitated and, clogs the filter. Series V. These experiments were made in order t o ascertain what effect the time of previous heating of t h e diastase solution has on the amount of d-glucose formed. The diastase used was prepared from malt germinated by ourselves in the laboratory (see Ling and Davis, Zoc. cit., p. 484). Three portions of the diastase, amounting each to 0.3 gram, were restricted at 64' for 1 hour, 2 hours, aud 4 hours respectively.They were then each added t o three 15 gram portions of starch made into paste with 300 C.C. of water, the resulting solutions being placed in a thermostat at 53'. The following analytical results were obtained :POTATO STARCH PASTE. 2 3 Time of Time of cy.ga restriction. hydrolysis. (corr. 1. , ['ID 3'93' 1 hour 117 hours 4'291 150.i" 2 hours 117 ,, 4 -430 154.2 4 7 7 115 ) ) 4.1 59 151.4 78'9 ~ 6.5 7'4 74 -1 7'3 73.1 These results indicate that, when the time of restricting a t 66' is more than an hour, there is no substantial increase in the amount of d-glucose formed. Swies VI. A freshly prepared active sample of diastase from Odessa malt was employed ; 0.3 gram was used in each experiment and 15 grams of starch.The temperature of restriction was 65-66'. Tiiiie of restriction i n liours. u Time of hydrolysis ~ in hours. 1 17 I 41 I 113 209 17 41 113 209 4.072 4.147 4.345 4'384 4.081 I 4-175 4.053 i 4.278 I - 152'ii" ~ 154'0" 144'8 145.5 14.5 *;; 146'2 145':; I 147'8 :: 1 113 209 I I 17 41 113 209 74 '4 72.6 57% 85'1 87.1 j 83.5 85.4 1 83.4 4'2 ' 4.0 8 '1 8.0 9.9 9.2 4'3 7.6 2. I 4 . 3.873 3.988 4 034 4.079 3'8SO ~ 3'965 3.869 3.378 I 154.7" 147'8 147 '3 148'8 158.4" 153.2 153.3 153.1 66.3 73.4 73.4 74.5 nil 2.2 8.0 7.5 * In this experiment, a distinct amount of mitter insoluble in boiling water was obtained, but it was amorphous, and therefore not glucosazone. It is seen by these experiments that for a restriction temperature of 66" the maximum amount of d-glucose is formed when the period of26 DAVIS AND LING: ACTION OF MALT DIASTASE ON heating was hal€ an hour, It is interesting to note the marked diminution in tbe percentage of d-glucose, after the hydrolysis has pro- ceeded for 209 hours, as compared with the values in the preceding columns (1 13 hours).This is possibly due t o the condensing action of the enzyme. Series VII. The same diastase was used as in the last series, but the tempera- ture of restriction was 69'. The temperature of hydrolysis was 5'. Time of restriction i n hours. Time of hydrolysis 0.25. in hours, I I I 1 7 ;; I 137 l i 41 65 137 17 41 65 137 r I 17 1 41 65 137 3.924 3'811 3'724 3.643 154 -6' 141.4 139.1 139'4 __. - - 91'1 0.5. 4'044 3.876 3'923 3.810 1.4'093 3.950 3.894 - 2. 3.980 3'919 3'852 3-882 157.1" 144-3 143.1 143'0 161'5" 150'8 148'5 I 165'2" 157'7 150'7 151 *2 96'6 89 3 - 96 '6 90% ~ 81'4 7.8 11'6 11'8 10'8 8 -7 11-2 10.6 10.6 5 '4 8'2 7.2 * 7.9 - 69'4 73'3 75'2 1.1 7.5 6.5 1 *5 * The exact concentration of this solution was not deteriniued, and the figure G,.,,=7*2 is calculated on the concentration found for the solution after 41 hours. Seyies VIII. I n these experiments, the temperature of restriction was raised t o 70-5', but the time of heating was reduced. Temperature of hydrolysis = 5'.POTATO STARCH PASTE. 27 I i Time of hydrolysis in hours. I Time of restriction in minutes. _c____ __ _ _ _ ~ 5 . I 17. 1 30. 1 120. 4.079 4.121 1 3985 4 016 1 4'063 ' 4.005 3'995 4.061 I 3.926 19 43 67 4.027 4.009 3.969 166%" 162.3 162'2 19 8.9 ' 8.1 5'8 4'7 10-0 7.3 1 3-9 6'8 1 7-9 3'7 Series IX.A . 0.3 gram of diastase restricted at 61.5" for 45 minutes and added at 55" to 20 grams of starch made into paste with 400 C.C. of water. 1 4 335 4'255 4'271 ' I 148.2" ' 13S.7" ' 137% 1 51'2 1 96'3 1 ~__- I 66 - I, B. The amounts of diastase shown in t h e following table were88 - DAVIS AND LING ACTION OF MALT DIASTASE ON hydrolysis iu hours. - 18 4 2 66 Time of restriction in minutes. 5. 4.271 4.219 4.120 18 42 66 145'6" 136.1 135'3 1 5. I-_____ 4.107 L.082 4,077 15. _ _ 4.041 4.009 4.000 141.8" i 133'4 I 133'1 I 1453" 135'6 135.5 18 84.6 42 I 98.7 66 99.8 90.1 102.4 101.6 87.4 101.9 100.2 18 7.7 - 9 *7 66 10.7 8.9 9.4 9.6 11.4 1 11'0 C. One gram of diastasemas dissolved in water,nnd the solution, which was raised t o a temperature of 7 3 O in 5 minutes and maintained a t 73-73-5' for an additional 5 minutes, was then added to 20 grams of starch made into paste with 400 C.C.of water. Series X. I n this series, solutions containing the weights of diastase given in the table were restricted at the temperatures mentioned, rapidly cooled, and added to paste prepared from 20 grams of starch and 400 C.C. of water at 55" in each case. The hydrolysis was allowed to proceed for 66 hours.POTATO STARCH PASTE. 29 Amount of Temperature of diastase. restriction. 0.3 61'. 5 9 .5---60' 0.3 9 , 63" 0.3 :, 6 2' 1.0 97 76*5---77'5' I n the last experiment, waker, a t with the dry diastase, the mixture 76.5".The solution was raised to Time of restriction. G,. 43' 1 hour, 0.5 0-25 ,, 0.4 0-75 ,, 0.5 5 min. 3.9 a temperature of 78", was mixed having initially a tempereture of 77.5' in 5 minutes, after which i t was rapidly cooled and added to the starch paste. Gomlusions. The effect of heating a solution of diastase is t o cause a weakening of the action of the enzyme, and also t o produce an alteration in the diastase molecule. The latter effect, which is the more important in the foregoing experiments, is a permanent one, for diastase solution which has been heated above 55' retains its altered properties when reprecipitated from a solution by alcohol and allowed to act on starch paste a t a temperature of 55" or below, Tho alteration of the diastase appears t o commence when a solution is heated below 60°, although a complete change is not effected, inas- much as the amount, of d-glucose formed by the action on starch paste of diastase, previously heated for an hour at this temperature, is small. As the temperature of restriction is increased, the amount of d-glucose formed by the action of the enzyme on starch is augmented, and the maximum amount of d-glucose is produced by diastase which has been previously heated i n solution at 68-70". Above this temperature, the weakening of the enzyme is so rapid that a much larger quantity of it has to be employed in order t o attain the stage of the reaction a t which d-glucose appears, especially if the heating is prolonged ; never- theless, this hexose is formed by diastase restricted a t temperatures up to 78', and probably above this. I t has invariably been observed that, when the solution is kept a t the temperature of hydrolysis, usually 55*, after the maximum amount of d-glucose has been formed, this sugar diminishes i n amount, and the occurrence of this apparently condensing action of the enzyme may probably explain the occasional failure t o detect d-glucose among the products of hydrolysis (see Series I and 11). I n any case, the maxi- mum amount of d-glucose formed does not exceed 1 2 per cent. of the total hydrolytic products. 74, GREAT TOWER STREET, LONDON, E. C. - ----