SUMMARIES OF THE HISTOLOGIC CHARACTERS, ET< 



285 



hybrids from the same cross (see Brunsdonna) ; but 

 here again in the final summing up there is usually 

 found to be a distincl majority of the reactions leaning 

 to one or the other parent. It i- unfortunate thai very 

 frequently the data have uoi been recorded in accord 

 ance with the plan adopted at the outstart of the re 

 so as i" leave no doubt in each character or character 

 phase of the parental relationships of the hybrid, such as 

 was pursued in making the quantitative determi ation . 

 Owing to this defect it is necessary to present these 

 summaries in a modified tabular form, and h ith the \ iew 

 particularly of showing the fluctuating relationships of 

 the hybrids to the parents. In the preparation of the 

 tables that follow (Tallies c I to C L7), the properties of 

 the hybrids in their parental relationships have been 

 considered collectively in designations or groups that 

 arc indicated by the divisions of the tables, those of form 

 being taken as one designation, those with a given rea 

 gent as one designation, and so on. The plus sign is to 

 he interpreted a- meaning that in the final summing up 

 of the data of each designation the hybrid in its unit- 

 character and unit-character-phase hears, on the whole, 

 a closer relationship to the parent indicated at the head 

 of tin' column. The minus sign is, of course, the nega- 

 tive correlative of the former; while the plus-minus sign 

 indii ates that the hybrid resembles in decree one as much 

 as the other parent. In the hist column the terms excess 

 and deficit mean that a unit-character or unit-character- 

 phase is developed in excess or deficit of parental ex- 

 tremes: individual means that a unit-character or unit- 

 character-phase has been discovered in the hybrid that 

 was not observed in either parent. 



Certain apparently minor peculiarities have been dis- 

 regarded in this tabulation. In sonic instances it is 

 entirely arbitrary whether we regard a given property as 

 being developed in excess or deficit of parental extremes. 

 Thus, if the grains of the hybrid be more irregular, or 

 the resistance to reagents greater, than those of the 

 parents, are we to look upon the difference as being an 

 expression of increased or decreased development? Ten- 

 tatively, such differences have been taken as represent- 

 ing increased development; and, if there be less irregu- 

 larity or less resistance, the opposite. It is obvious that 

 these tables indicate merely very grossly certain promi- 

 nent phases of hybrid and parental relationships, and that 

 the context must be studied therewith in order that the 



qualitative and quanl 



h parent can prop, rl be in In 



of tabic- that follow, the 

 and v - 6 are used as sc.\ designal ion 



t, nearer the pollen parent, and equally 

 ively. The symbol s in 'I 

 !■', 1 to 50, and II, 1 to 2G, ind 

 are too fa t oi I ition, 



it becausi I 

 tini at different 



or sufficiently definite inclina The 



data of the quantitatii the 



various table- of the reai tion it 1 by 



;l "' i tagi ■ tan b gelatinizi 



ntervals that const ituti the thin 



mary in Chapter III, ami also tabulated in modified ar- 

 ran ement in Sect ion l of this chapter. T i have 



also be. -ii presented in the IV. 



It is important to note that in the studii - of the quali- 

 tative reactions the reagent eld ed varied somewhat 

 in number and kind in tl 



hybrids, and thai in the formulation of tl tables the 

 quantitative reactions givi □ are limited ti of the 



reagents used to elicit the qualitat 

 m the summing up in these tabl< 



the reactions of the hybrids I I of the parents there 



may seem to In' some discrepancies v. hen the 6 

 compared with those of Tables E, l to 22, I . 

 and II. 1 to 26. for instance, in the quantit 

 tions of Brunsdi ■ ha it will be noted that 



of the 8 reactions with the chemi - like 



that of the seed parent, pollen pa r both pai 



1 is intermediate, I is higher than that of eithi 

 and t'i are lower than those of either parent. When, 

 however, all of the 21 reaction are summed up it is 

 found (TaMc I'. I ) that I are the same as 

 parents, none the same as those of the pollei ; 



the same as those of both parent-. 5 intermedial 

 higher than those of the parents, i >ver than 1 



of the parents. 



The limited quantitative data given in Table- C 1 

 to ( ' 11 are mainly for i 



reactions with the same reagents, the data of this kind 

 being tabulated in full in table- E, I-', and 11. Limited 

 comment only i.- necessary in explaining tin- series of 

 tables. 



(a) Brunsdonna sanderoe alba (sami parentage as following hy 



Table C 1. — Brunsdonna sanderm alba. 



