ki;a( th>n-intensities with each agent am- reagent. 



17:; 



: i mi!- if fallacy by increasing the number and 

 changing the concentration of the rea -I modify- 



ing the standard of values in accordance with the ab 

 Mr here used. Notwithstanding the crudities i 

 methods adopted ami the fallacies introduced in the 

 formulation of the composite charts in tip' former 

 memoir the following was rendered apparent: Thai I ie 

 reactions oi members of a genus coi titute a well-defined 

 group, the mean of the character-values constituting a 

 distinct generic type, tin- type tending to he similar to 

 th.' types of vi l\ related genera ami dissimilar 



to tlir types >>( distantly related or unrelated genera; 

 that tho reactions ol diffen m species of a genus yield 

 curves that tend to ely in conformity with the 



generic type of curve, bui when there are representatives 

 ul' subgenera or similar generic subdivisions there may 

 ho departures or aberrations from this generic typi 

 that there may he as many subgeneric or urri ni| > types as 

 there arc subgenera or subgeneric groups; that tho reac- 

 tions of variel ies of a species \ ield curves thai very closely 

 correspond with those of the species ; ami that tho generic, 

 subgeneric, ami species differentiations arc in general 

 in close accord with established botanical data. The re- 

 sults of tho preseni research are in harmony with those 

 of the preceding investigation, hut some unexpected 

 variations have been found, especially in tho extent of 

 certain generic and subgeneric differentiations which will 

 i ferred to here with sufficient detail. 



Taking up firs) those genera which arc best repre- 

 I by species and varieties, hut in which there are 

 not included subgeneric or similar generic gToup repre- 

 sentatives, such a- Hippeastrum (Charts E2, E 3, and 

 I! l i. Nerine (Charts E 10, E 11, and E 12), Narcissus 

 (Charts E 13 to E 24, inclusive), and Lillian (Chart- 

 E25 to E 29, inclusive), it will he apparent upon 

 even superficial examination that the starches of the 

 varieties or species, or of both varieties and species, of 

 each genus have curves that are in general very similar 

 in form and that the type form of the curve in each genus 

 is different from that of any other, and so markedly 

 so that the curves of the members of one genus could 

 not he confounded with those of another any more than 

 en uld the plants themselves. It will also he noted that 

 when He starches are from very closely related plants, 

 as in the Hippeastrwms, the curves arc very closely alike, 

 while in Nerine and Narcissus, res ei tively, "here there 

 arc instances of both botanical closeness and separai 

 the variation- from the mean or the generic type of 

 curve tend to he more and more marked a- the repre- 

 sentatives of the genus are botanically farther separated. 

 The curves of Lilium, while yielding a generic type very 

 different from the Hippeastrum, Nerine, and Narcissus 



- -. arc of little usefulness in the differentiation of 

 the various members of the genus represented because 

 of the very rapid gelatinization of the starches with 

 nearly all of tl reagents, hi order to satisfactorily 

 differentiate these starches reagent- of such modified 

 strengths must he used a- will render gelatinization very 

 much less rapid, and probably additional reagents may 

 he necessary. In other genera studied, where there arc 

 only the two parental and the hybrid representatives of 

 the | in Gladiolus i < ! hari E 34), Tritonia 



(Chad E 35), Eichardia (Chart !•'. Hi). Musa (Chart 

 E ID. Phaius (Chart E 12), Miltonia (Chart E 13), 



i 'ymbidium ( < hart E 44), 

 \mI1 be found, although it 



related gem ra, I he curves are .-■, much al indi- 



liffcrenl spei ie rather than dim • ra. There 



is also much i the A marylli 



Phaius charts which represent very w 

 genera, hut iln i culiarity will i ed to 



particularly later on. lathe An < igia reac- 



( ( 'hart El), « bi 

 tion, the curves are quite different. 



When genera an 

 generic groups, as in Hainan I E 6), < V 



(Charts E 7, E 8, and E 9), Iris (I harl I E 31, 



E 32, and i: 33), and Begonia (Charl E3G), the curves 

 of the subgeneric representatives may d I onlj 



markedly hut to even a much more marked degree than 

 the curves of diffi i generally of the 



family — a most curious and as yet inexplicable 

 nomenon. In Hamanthus the curve of //. puniceus 

 variant in comparison with //. katherina, If. 



magnificus, and both hybrids thai it seems that this 

 en- musl be separated botanical far from 



the other two to be regarded a- belonging to a diffi 

 subgenus, although this differentiation may nol have 

 recognized by the systematist. In Crinum the curvi o 

 the representatives of the hardy and tender 

 moorei and C. longifolium, hard} ;C. ilanicum, tender) 

 differ so markedly as to suggesl members of difl 

 genera. In Iris, in the first three gets 

 E 31, and E 32 i, the read ions of rhyzomatom I n 

 represented, and it will be seen that all urves 



conform closely to a common typi : bui in the Court! 

 I ('hart E 33) the reaction- are of i ! form-, all 



three curves conform with great closi I immon 



type, and they all differ materially from the rhyzomatous 

 type, and in fact so different are they that they would 

 certainly not in the preseni stages of the ini 

 be recognized as belonging to tb n *enus. In Be- 

 gonia there is found an even more remarkable instance 

 of subgeneric differential ion in the curves of the tubi 

 and semituberous forms, the fort ted 



by four garden varieties and the latter by 11. socotrana, 

 a very exceptional and isolated species of the genus. 

 Comparing the curves of these chart- (Chart- E 36 to 

 E39) it will be seen that the curve- pf the in 

 form- are in close conformity to a common type, while 

 the curve of />'. socotrana is so very unlike the curvi 

 the former in a large number of the reactions with the 

 chemical reagent- as to suggesl anything but generic 

 relationship to the tuberous forms. Unfortunately, the 

 number of reaction- of the 1 re with a -in. 



eeption very limited, but the curve of the reactions of />'. 

 single crimson scarlet (Chart E 36) can witl 

 safety be taken as very closely typifying the cun 

 tl thers. 



The Amaryllis and Phaius curve- (Charts E] and 

 E 12), while representing wholly unrelated and widely 

 separated genera, give the impression of < i losely 



related genera or even of species of a genus: in fact, the 

 nblance is much closer than that of related orenera 

 hererepr ted, as, for instance, of A maryl Iruns- 



vigia (Chart E I ). of Phaius and Miltonia I charts E l\! 

 and E 13), or of Phaius and Cymbidium E I? 



and E II). While there is some re 



