158 



EEACTION-INTENSITIES OP STARCHES. 



of the tender Crinuin, so that in every chart the curves 

 of these three species are V-shaped, and the first segment 

 of the V is longer than the second, the difference in 

 length varying with the different reagents. In 7m the 

 first three specimens are definitely differentiated from 

 the fourth in most of the charts by the distinctly lower 

 reactivities of the former, the exceptions being in the 

 reactions of chloral hydrate, chromic acid, sulphuric acid, 

 potassium sulphocyanate, potassium sulphide, and so- 

 dium salicylate (in the chloral-hydrate and potassium- 

 sulphide reactions those of the former are the higher). 

 In other words, in only 4 of the 21 reactions is there not 

 a definite separation of the first three from the fourth. 

 In Begonia the differentiation is not only very marked, 

 but also in certain respects extraordinary : B. socotrana 

 is a very exceptional form of the genus, is semituberous, 

 and is botanically quite different from the tuberous Be- 

 gonia single crimson scarlet. The starches of the two 

 plants in histologic and polariscopic characters, qualita- 

 tive reactions with various reagents, are alike in many 

 respects and very dissimilar in others, so that each ex- 

 hibits certain striking and distinctive characteristics (see 

 Chapters III and V, and Part II, Chapter VIII). These 

 peculiarities together with the remarkable differences in 

 their reaction-intensities constitute one of the excep- 

 tionally interesting findings of this research. 



The curves of the reactions of the four tuberous Be- 

 gonias (Charts E 36, E 37, E 38, and B 39) tend to be 

 as much in accord as should be expected in plants that 

 have such a botanical relationship, but the curve of B. 

 socotrana (Chart E 36) appears definitely to be vagrant 

 in nearly all of the reactions. The four hybrids incline, 

 on the whole, to an obviously closer relationship to the 

 tuberous parents than to B. socotrana. Examinations 

 of the curves of the preceding charts (Charts B 11 et 

 seq.) will show that: With chloral hydrate there is 

 definite but not marked differentiation, 99 per cent of the 

 total starch of B. single crimson scarlet being gelatinized 

 in 10 minutes and 95 per cent of the starch of B. soco- 

 trana in 15 minutes. With chromic acid there is 98 per 

 cent in 15 minutes and 92 per cent in 60 minutes, re- 

 spectively, a wide difference. With pyrogallic acid, 95 

 per cent in 45 minutes and only 0.5 per cent, or almost 

 nothing, in 60 minutes, giving a much wider difference 

 than with the preceding reagent. With sulphuric acid 

 a practically complete gelatinization occurs in both 

 starches in a minute, while with hydrochloric and nitric 

 acids with the starch of the first plant there is immediate 

 gelatinization with both reagents ; and with B. socotrana 

 with the hydrochloric acid there is 45 per cent in 45 

 minutes, and with nitric acid only 12 per cent in 60 

 minutes. With potassium hydroxide there is an almost 

 instantaneous gelatinization of both starches. With po- 

 tassium iodide there is practically complete gelatiniza- 

 tion of one in 30 seconds, while with the other there is 

 almost no detectable effect, only about 1 per cent being 

 gelatinized in 60 minutes — almost the absolute extremes 

 of reaction-intensity. With potassium sulphocyanate pe- 

 culiarities are elicited that are almost identical with those 

 of the last reagents, the only difference being a some- 

 what larger percentage of starch of B. socotrana gelati- 

 nized in 60 minutes — ^here 18 per cent. With potassium 

 sulphide the differences between the reactions of two 



starches is positive, complete gelatinization occurring in 

 the starch of B. single crimson scarlet in 15 seconds and 99 

 per cent in the case of B. socotrana in 5 minutes. With 

 nearly all of the remaining reagents (including sodium 

 hydroxide, sodium sulphide, calcium nitrate, uranium 

 nitrate, strontium nitrate, copper nitrate and cupric 

 chloride) gelatinization of the starch of B. single crim- 

 son scarlet is with each reagent complete within 2 min- 

 utes, while with the starch of B. socotrana it varies from 

 0.5 per cent to 84 per cent in 60 minutes (with two 

 reagents there was 84 per cent, with one 35 per cent, with 

 one 9 per cent, with one 1 per cent, and with two 0.5 

 per cent). With sodium salicylate the figures for the 

 first starch are 97 per cent in 3 minutes, and for the sec- 

 ond 99 per cent in 10 minutes. With cobalt nitrate the 

 figures for first are 66 per cent in 60 minutes (the low- 

 est record for this starch with any of the reagents), and 

 for the second 0.5 per cent in 60 minutes. With mer- 

 curic chloride the first starch shows a gelatinization of 

 96 per cent in 15 minutes, and the second 0.5 per cent 

 in 60 minutes. The extraordinary differences exhibited 

 by these starches are at present inexplicable, and they 

 open a field of most interesting and promising research 

 of the most fundamental character. 



Inversion and Reversion of Reaction-intensities. — 

 The inversion and reversion of the reaction-intensities 

 of different starches with different pairs of reagents is 

 also a feature of exceptional interest and of pre-eminent 

 importance in proof of the existence of starches from 

 different plant sources being in stereoisomeric forms. 

 It is obvious, as before stated, that if we were dealing 

 with starches that differ from each other because merely 

 of differences in density, reaction, impurities, percentage 

 of water, or varying proportions of several modifications 

 of starch in the form of mechanical mixtures, the two 

 curves would be alike or one would always be above the 

 other, the distance, however, varying in relationship to 

 the rapidity of reaction, the slower the reaction the 

 greater probably the tendency in general to separate. It 

 has been repeatedly noted that inversion and reversion of 

 the curves is not limited to the distinction of genera, 

 although it is more apt to be associated with genera, and 

 next in order with subgeneric groups, and next with 

 species. In other words, if with any two reagents a 

 member of a given genus will exhibit a greater reactivity 

 with one than the other reagent the same peculiarity 

 will probably be found with all other members of the 

 genus unless there are definite subgeneric divisions of the 

 genus, under which conditions the subgeneric divisions 

 may be as distinctly differentiated as may be genera by 

 inversion or reversion of the reaction-intensities. 



Sometimes species of a genus which are not recognized 

 as belonging to subgeneric groups may exhibit inversion 

 or reversion in their reactivities in relation to the reac- 

 tivities of the other species, as has been found, for in- 

 stance, in Nerine. These inversions and reversions are, 

 as a rule, not so apt to occur with reagents of a similar 

 as of a dissimilar character. Moreover, the points at 

 which inversions and reversions of the curves of any pair 

 of reagents occur may be the same or different from those 

 at which inversions and reversions of another pair 

 occur — ^that is, two genera or representatives of two 

 subgeneric divisions, or two species of a genus, may be 



