1,1 ACTION-INTENSITIES WITH EACH ACENT AND REAM HI 



1 II 



chiefly becau e of the comparative less reactivity of the 

 firs! Bet of parents and hybrid tliaa of the othi i 

 In Chart A. 6 (chloral-hydrate reactions) there is 

 marked exten ion of the maximal and minimal limits of 

 the reactions owing to the prolongation of 1 of the 11 

 lines, so thai the group is nothing like so distim tly in 

 dividualized as in the 11 charts referred to wherein the 

 maxima and minima are close. In Charts A. 9, All, A 12, 

 A ll, A 15, and A21 (nitric acid, hydrochloric acid, 

 potassium bydroxide, potassium sulphocyanate, potas- 

 sium sulphide, and strontium nitrate) there is a well- 

 marked separation of the first from the second and third 

 Bets, the latter showing about the same, and the former 

 distinctly higher reaction-intensities. Such peculiarities 

 arc found to he common among the other genera where 

 a number of sets of parents and hybrids are included, 

 from which it is obvious that where a genus is represented 

 by a single such set the maximum, minimum, and mean 

 reactive-intensities are to be taken merely tentatively as 

 representing the generic standards. 



This statement finds immediate application to a num- 

 ber of generic groups represented in these charts, includ- 

 ing Amaryllis-brunsvigia (bigeneric), Gladiolus, Trito- 

 nia. Richardia, Mum, Phaius, Miltonia, and Cymbidium. 

 The maximum, minimum, and average values ditfer not 

 only in the case of differeni sets of parents and hybrids 

 of the same genus, but also of the members of the same 

 set with different reagents. Thus, in Nerine, in Charts 

 A 8 and A 11 (pyrogallic-acid and sodium-sulphide reac- 

 tions) and in certain other charts, the maxima, minima. 

 and averages for all of the species and hybrids are prac- 

 tically absolutely the same, but in Charts A 11 and A 1 I 

 (hydrochloric-acid and potassium-sulphocyanate reac- 

 tions) and in others, all three are different in all three 

 sets of starches. Finally, generic grouping may seem- 

 ingly be set aside in some instances by wide differences 

 in the reaction-intensities of one or more sets included 

 in the genus group. This is well illustrated in Crinum, 

 Iris, and Begonia in Chart A 9 (nitric-acid reactions). 

 The species of Crinum studied in this research are divisi- 

 ble into two horticultural groups, which are distinguished 

 as tender and hardy, the starch of the former being char- 

 ai terized by generally low reactivities and those of the 

 latter by generally high reactivities, the differences being 

 so marked that it is necessary to recognize in these 

 starches two distinct subgeneric groups. Such differ- 

 ences are well shown in other charts, such as Chart- A 8, 

 A 10, A 11, and A 12, but then' is an entire absence of 

 such distinction in Charts \ 6, A 7, A 15, A 19, A 23. 

 A 23, A 25, and others. In fact, in several of the latter 

 the differences are so slight as to suggest \er\ closelj 

 related member- of the genus. In Iris there is a very 

 conspicuous example of subgeneric grouping: In Charts 

 A 5, A 6, A ", A I", and A r> the reaction-intensities of 

 the members of all four sets are nearly the same or do not 

 differ to a marked degree; but in A 8, A 9, A 11. A 12, 

 A 13, A 1 1. A 16, A 17, A 18, A If), A 20, A 51, A 22, 

 A 23, A 24, A 25, and A 26 there is a well-marked group- 

 ing, the first three sets constituting one group and the 

 last set another group. 



With the exception of Charts A C, and A 18 the first 

 group is characterized by lower reaction intensities, 

 which with rare exceptions tend to be very close in all 



sets, thus forming a vi tincl group. While 



in ChaM A 6 and A18 tl m.-. there 



is a reversal of the reaction-intensities, the | 



showing less reactivity than the second group. E en 



■ i- Begonia: In Chart A 'J there 

 obvious differentiation of anj of the gets of member 

 set, but in Chart A 6 there appears a very conspicuous 

 differentiation in the comparativi slowni of tic- /•'. 

 socotrana reaction; and in all other chart-, with four 

 exceptions, the length of the line i- accentuated in vary- 

 ing degree, thus markedly i haracterizing the si 

 group. This seemingly aberrant reaction-inti 

 this exceptional species gives a peculiar generic pi 

 and means, as in the instances of ' rinum and Iris, two 

 generic type-. 



The correspondence of the grouping of the reaction- 

 intensities of starches in accordani neral with gen- 

 era is usually quite evident, this being no1 only more 

 marked with some than with other agents and reagent . 

 as stated, but also more marked with so with 

 other groups. A given group may stand out very con- 

 spicuously in one chart, but not in another, or even not 

 be differentiated from adjoining groups, yet be more or 

 less distinctly differentiated from the same groups in 

 other charts. For instance, in Chart A 10 (sulphuric- 

 acid reactions), taking the genera represented by Nerine, 

 Narcissus, Lilium, Iris, Gladiolus, and Tritonia, it will 

 be seen that with the exception of Gladiolus there is no 

 differentiation of the reaction-values that even -u - 

 that the records are those pertaining to differeni genera ; 

 in fact, they are so nearly alike 1 as to indicate that the 

 several groups belong to a single genus. The Gladiolus 

 reactions take place with comparative slowness, which 

 distinctly differentiates this genus from the five other 

 genera. In Chart All (hydrochloric-acid reactions) 

 Lilium stands very distinctly apart from the other five 

 genera : Nerine and Narcissus are not differentiated 

 from each other, but they differ from Lilium, Iris, Gladi- 

 olus, and Tritonia. 



It will be seen that three of the four set- of Irids 

 are practically alike and markedly different from the 

 fourth set, showing what marked differences may be 

 exhibited by members of subgenera or of similar divi-i- ins 

 of genera. In Chart A 12 (potassium-hydroxide reac- 

 tions) the picture is radically changed in a number of 

 particulars: Lilium remains conspicuous as before; Ne- 

 rine and Narcissus are very definitely grouped, the lines 

 of the former being very short and those of the latter 

 quite long; Iris differs but little, as a whole, from the 

 preceding chart ; and in both Gladiolus and Tritonia the 

 lines are prolonged ami about the same, giving no differ- 

 entiation between these two genera. In Chart A 13 

 (potassium-iodide reactions) the picture again differs: 

 Lilium is about the same; the Nerine lines are very con- 

 siderably prolonged and markedly exceed the length of 

 the Narcissus lines which are slightly shortened in com- 

 parison with the length in the preceding chart, thus show- 

 ing a marked reversal of the quantitative relationships. 

 The Narcissus lines and those of the first three sets of 

 Irids are about the same, whereas in the preceding chart 

 the latter are. on the whole, distinctly shorter: and 

 Gladiolus and Tritonia are about the same, but longer 

 than the Narcissus and Iris line-, and shorter than the 



