302 



SUMMARIES OP THE HISTOLOGIC CHARACTERS, ETC. 



Phaius. (Table C 14.) 



With the exception of the character of the hiluni 

 and the reaction with iodine the hybrid in its histologic 

 properties and qualitative reactions is closer to the seed 

 parent. Excess of development is noted once; deficit 

 and individuality not at all. 



The quantitative reactions are very variable in their 

 parental relationships, exhibiting sameness in relation to 

 one parent or the other or both parents, intermediateness, 

 and excess or deficit in relation to parental extremes, as 

 the case may be. 



MiLTONiA. (Table C 15.) 



Except in the eccentricity of the hilum and size of 

 the grains all of the designations of this hybrid incline 

 toward the seed parent. 



The qualitative reactions while variable in their 

 parental relationships tend with one exception to the 

 seed parent, but in none to the pollen parent. 



Ctmbidium. (Table C 16.) 



The hybrid bears a closer relationship to the seed 

 parent in all of the histologic and qualitative designa- 

 tions with the exception of eccentricity of the hilum and 

 of ratio of length to breadth of the grains. 



In the quantitative reactions the inclination is, with 

 one exception, to lower reactivity than in either parent, 

 the hybrid being in the latter reactions lower than in 

 either parent but as close to one as to the other parent. 

 The leaning is generally very doubtful because of the 

 great rapidity of the reactions. 



Calanthb. (Table C 17.) 



In 0. veitchii two-thirds of the designations incline 

 to the seed parent. In form most of the grains are more 

 like those of C. rosea, and only some like those of the 

 other parent. In hilum and lamellae the hybrid is close 

 to the seed parent, but in size closer to the other parent. 

 In the polarization figure, selenite reaction, and iodine 

 reaction it is closer to the seed parent. In the qualita- 

 tive reactions with chloral hydrate, potassium hydroxide 

 and sodiimi salicylate it is closer to the pollen parent; 

 but in those with chromic acid and hydrochloric acid it 

 is closer to the seed parent. In the quantitative reactions 

 throughout the hybrid is the same as or closer to the 

 seed parent. 



In C. iryan the designations are about equally divided 

 in their parental closeness. In form some of the grains 

 are more like those of the seed parent, but most are like 

 those of the pollen parent — ^the reverse of what was 

 recorded in the other hybrid (in this set the seed parent 

 is the same as the pollen parent in the preceding set). 

 There is in this hybrid in comparison with the other hy- 

 brids reversal of the relations of the hilum and lamellae 

 to the parents, and there is a splitting of the characters 

 pertaining to size — ^the grains in ratio of length to 

 breadth being closer to the seed parent, but in size gener- 

 ally closer to the pollen parent. While the polariseopic 

 figure and selenite reaction are in comparison with the 

 foregoingi hybrid reversed, the iodine reaction remains 

 closer to the seed parent. The qualitative reactions like- 

 wise show curious differences. Here the chloral hydrate, 

 chromic acid, and sodium salicylate reactions are closer 

 to the seed parent, while the hydrochloric acid and po- 

 tassium hydroxide reactions are closer to the pollen parent 



(the reactions of chloral hydrate, hydrochloric acid, and 

 sodium salicylate being reversed, but those of chromic 

 acid and potassium hydroxide remaining the same in 

 comparison with those of 0. veitchii). 



The quantitative reactions exhibit a tendency to mid- 

 intermediateness, and otherwise mostly to closeness to 

 the pollen parent. In only one of the seven quantitative 

 designations is there manifest greater closeness to the 

 seed parent than to the pollen parent. 



Histologic Peopeeties of Staeches of Hybeids 

 IN Relation to those of the Paeents. 



In the preceding section, in the consideration of the 

 peculiarities of each starch, reference was made to the 

 remarkable shifting of the various histologic characters 

 in their parental relationships. These peculiarities are 

 of exceptional interest and significance, and they have 

 been presented for the most part in a succinct form in 

 Table D. One would not unnaturally be led to the 

 conclusion that if the grains of the hybrid are closely 

 like those of the seed parent or the pollen parent in form, 

 lamellffi, and size, the same would hold good for the 

 hilum, but such may in fact be far from the case. 

 Moreover, not only may there be different parental rela- 

 tionships of the hybrid starch in form, hilum, lamellae, 

 and size, but there may also be a splitting of characters 

 in each of these designations, so that in a certain respect 

 the hilum, for instance, may be close in its relationship 

 to one parent, but in another respect equally as close to 

 the other parent. In other words, not only are form, 

 hilum, lamellae, and size independent characters that 

 may be modified in the starch of any hybrid in their 

 parental relations in like or unlike directions, but each 

 may be split into a variable number of components which 

 in like manner may swing to one or the other parent in 

 an absolutely unpredictable and inexplicable way. It is 

 unfortunate that in making the laboratory records the 

 data pertaining to variations in form were not so syste- 

 matically made as to make it possible to present in a 

 consistent way the splitting of properties such as was 

 recorded in the properties of the hilum, lamellae, and 

 size, especially of the two former. Sufficient data were 

 accumulated to show that such splitting is a common 

 phenomenon, as, for instance, where it has been found 

 that the hybrid is close to one parent in the characters 

 and numbers of compound grains, but close to the other 

 parent in the characters and numbers of the aggregates ; 

 where a certain type of compoimd grain or aggregate is 

 closer to that of one parent, but another type closer to 

 that of the other ; where the kinds of irregularity of the 

 grains incline to one parent, but the frequency of irregu- 

 larity to the other, etc. Similarly, only little analytic 

 attention was given to the pecidiarities of sizes, but 

 enough to show that a splitting of characters must be 

 quite common. On the other hand, the records of the 

 peculiarities of the hilum and lamella, while capable 

 of much and important extension, are rich in instances 

 of splitting. Taking several concrete examples for illus- 

 tration, we find that both Brunsdonna hybrids are 

 closer to the seed parent in form, hilum, and size, but 

 closer to the pollen parent in the form, arrangement, and 

 number of the lamellae. Hippeastrum titcm-cleonia is 

 closer to the seed parent in form and hilum ; but closer to 



