REACTION-INTENSITIES WITH EACH AGENT AND REAGENT. 



157 



socotrana. Additional comparisons of the data of these 

 charts will bring out many interesting facts. 



The potassium-sulphide and sodium-sulphide chart 

 (Chart B 34) bears in certain respects closer resem- 

 blances to the hydroxide chart (Chart B 33) than to the 

 acid chart (Chart B 15), and in other respects the re- 

 verse, thus indicating that the alteration of the hydrox- 

 ides into the sulphides has yielded reagents which give 

 rise to reactions that suggest the presence of both active 

 cations and anions, in contradistinction to the reactions 

 of the hydroxides and acids which are pre-eminently 

 cationic and anionic, respectively. These sulphide reac- 

 tions vary in intensity in both directions to almost the 

 extreme limits of the abscissa, from the extremely high 

 reactivities of potassium sulphide that are recorded in 

 Liliuiii,Begonia,nnd Phaius in which complete gelatiniza- 

 tion occurs in 2 minutes or less, to the extremely low 

 reactivities in Ilippeastrum, Hmnantlius, Crinum, etc., 

 where 5 per cent or less is gelatinized in 60 minutes. 

 The deviations of these curves from the acid and base 

 curves are much more marked than the variations of the 

 curves themselves, and the quantitative differences be- 

 tween the curves tend to be more marked and erratic, 

 and inversions to be more frequent, than in the acid 

 and base curves. In Nerine there occurs in the sulphide 

 curves, as in those of the hydroxide, an inversion, in 

 both charts the potassium salt is the stronger. In Iris 

 there is a marked separation of the curves, as was found 

 to be the case with one exception in the hydroxide reac- 

 tions; but in three of the starches there was no separa- 

 tion of the acid curves. In Begonia socotrana the curves 

 are less like those of the bases than of the acids, while 

 in Miltonia they stand apart from both base and acid 

 curves. The wide separation of the sulphide curves in 

 Amaryllis is very conspicuous in comparison with the 

 small separation of the base curves and the absence of 

 separation of the acid curves. Similar peculiarities 

 will be found in the reactions of these three pairs of 

 reagents with other starches. 



The potassium-iodide and potassium-sulphocyanate 

 reactions (Chart B35) bear, on the whole, far closer 

 resemblances to the hydroxide reactions than to the acid 

 or sulphide reactions. In contradistinction to the sul- 

 phides these reagents contain acid radicals that are 

 probably almost inert. Comparing this chart with the 

 base chart (Chart B 33), the most noticeable differences 

 will be found in the reactivities with Amaryllis, Bruns- 

 vigia, Hcemanthus puniceus, Nerine, Iris, Begonia, 

 Phaius, and Miltonia. Amaryllis and Brunsvigia each 

 exhibits practically no difference in the potassium-iodide 

 or potassium-sulphocyanate reactions, but Amaryllis and 

 Brunsvigia are differentiated from each other by both 

 reagents, both starches reacting more readily with po- 

 tassium iodide than with the other reagent. In Hceman- 

 thus puniceus, while these reagents do not differ in their 

 reactivities, potassium hydroxide yields a markedly dif- 

 ferent result from that of sodium hydroxide. In Nerine 

 reactivity with the iodide is very low and with the sul- 

 phocyanate low; while in the hydroxide reactions those 

 with potassium hydroxide are very high and those 

 with sodium hydroxide very low. In Iris the potas- 

 sium iodide reactions are very much lower in the first 

 three Irids and somewhat lower in the fourth ; while 



in the hydroxide reactions in two there are very marked 

 differences, in one no difference, and in another a 

 marked difference, the potassium reactions being the 

 lower when difference exists. In Begonia, the iodide 

 and sulphocyanatc reactions show very little difference, in 

 B. single crimson scarlet both reagents acting with great 

 intensity and in B. socotrana with great slowness, the 

 iodide being practically inert; while in the hydroxide 

 reactions both reagents act with great intensity with 

 B. single crimson scarlet, potassium hydroxide acts with 

 equal vigor, but sodium hydroxide with low intensity 

 with B. socotrana. In Phavus and Miltonia both the 

 iodide and the sulphocyanate show differences between 

 these genera and between the members of each genus, the 

 iodide being less active than the sulphocyauate. While in 

 both Phaius and Miltonia marked differences exist be- 

 tween the reaction-intensities of the iodide and the 

 sulphocyanate, there are comparatively small differences 

 between the intensities of the hydroxides. 



The curve of sodium salicylate (Chart B 36) stands 

 alone, as before stated, and therefore is not comparable, 

 as in the foregoing instances, with that of any other 

 reagent. 



Calcium nitrate and strontium nitrate (Chart B 37) 

 exhibit differences that are most pronounced in Bruns- 

 vigia, Crinum, Nerine, and Miltonia. The calcium curve 

 appears to correspond more particularly with the curves 

 of potassium iodide, potassium sulphocyanate, and so- 

 dium hydroxide; while the strontium curve appears to 

 be more closely related to the curves of uranium nitrate, 

 copper nitrate, cupric chloride, and mercuric chloride. 

 All of the latter curves appear to be very closely related 

 to a common type, which suggests that the reactions, in 

 so far as the latter depend upon the reagents, are due 

 essentially to differences in the basic ions or cations. 



Differentiation of Subgeneric Groups. There is 

 probably no feature of these charts more prominent or of 

 greater value in proof of the worth of the gelatinization 

 method in the differentiation of starches from different 

 sources than the constancy and definiteness in similar 

 and dissimilar directions of the differentiation of sub- 

 generic representatives. Hasmantlius l-atherincc and II. 

 puniceus are, from the standpoint of the systematist, at 

 most well-separated species, but from the results of this 

 research they are probably to be regarded as representa- 

 tives of well-defined subgeneric groups. Had this marked 

 subgeneric differentiation been indicated by the reac- 

 tions of a single or an occasional reagent it might natur- 

 ally be regarded as being accidental, but it is evident 

 throughout the charts of the reactions of the 21 reagents, 

 except the chloral-hydrate and sodium-salicylate reac- 

 tions. The one species is as definitely and widely differ- 

 entiated from the other as are genera in general, with 

 the exception only of the closely related Gladiolus and 

 Tritonia. While at the end of 60 minutes there is only 

 slight and questionable differentiation in the chloral- 

 hydrate reactions, and in the sodium-salicylate reactions 

 no differentiation, there are differences of importance 

 shown during the progress of the reactions (Charts D 106 

 and D118). The hardy and tender Crinums are with 

 every reagent markedly differentiated, but by some to a 

 better degree than by others. The abscissae of the two 

 hardy Crinums are in all of the reactions above those 



