374 



NOTES AND CONCLUSIONS. 



inasmuch as if in one species there be found, say, three 

 forms the same number will exist in other members of 

 the genus ; that the crystals of different genera differ as 

 definitely and specifically as those of crystalline groups 

 of mineral substances differ chemically and as generic 

 groups differ zoologically or botanically; and that by 

 means of peculiarities of the hemoglobins phylogenetic 

 relationships can be traced, as has been found in the case 

 of the bear and seal and other animals. 



Geneeal Conclusions deawn feom the Staech 

 Researches. 



The results of the hemoglobin and starch researches 

 are mutually confirmatory in support of the existence of 

 stereoisomeric forms of complex organic substances that 

 are specifically modified in relation to varieties, species, 

 subgenera, and genera, and that these s.pecificities indi- 

 cate corresponding peculiarities of the protoplasms in 

 which the substances are formed. The records of the 

 starch researches indicate: that each starch property is 

 an independent physico-chemical unit-character, and that 

 the unit-character represented by the property of gela- 

 tinizability may be manifested in an indefinite number 

 of quantitative and qualitative unit-character-phases, the 

 number varying with the form of starch and the number 

 of gelatinizing reagents employed ; that qualitative reac- 

 tions are as distinctive and important as the quantitative 

 reactions; that the reactions of different starches with 

 a given reagent vary within wide limits, and that those 

 of each starch vary with each reagent independently of 

 the variations of other starches; that the reactions of 

 varieties of a species very closely correspond to those of 

 the species and are in accord with botanical characters ; 

 that the reactions of members of a genus are in general 

 in close accord with taxonomic data and constitute a 

 generic type, the varieties and species tending to exhibit 

 closeness or separation in their relationships in close 

 accord with botanical peculiarities ; that when a genus is 

 represented by subgenera or other form of subgeneric 

 division (such as rhizomatous and tuberous plants, or 

 hardy and tender species, etc.), the reactions may exhibit 

 as many different groupings as there are subgeneric 

 divisions, and that these divisions may show very marked 

 differences, even more marked than what may be noted 

 in the case of closely related genera; that the reactions 

 of closely related genera tend to be similarly close; that 

 in hybrids any one of the six parent-phases (the same 

 as the seed parent, the same as the pollen parent, the 

 same as both parents, intermediate, higher than either 

 parent, and lower than either parent) can be developed 

 at will by the selection of the proper reagent ; that the 

 tendencies of different reagents to elicit in the hybrid 

 any given parent-phase varies with reagent and starch, 

 certain reagents tending to develop sameness to the seed 

 parent or to the pollen parent, etc., and a given reagent 

 may elicit one phase with one starch and another phase 

 with another starch, etc., so that by the selection of the 

 reagent any parent-phase can be developed in any given 

 starch ; that the starches of hybrids tend to show marked 

 closeness to the properties of the parental starches when 

 the parents are closely related, and to exhibit a tendency 

 to more and more divergence as the parents are more and 

 more distantly related, in some instances tending by 

 comparatively numerous intermediate characters to 



bridging the parental characters and in others to be par- 

 ticularly characterized by being very closely related to 

 one parent, or in others (by excess or deficit of develop- 

 ment) to be quite variant from the parental types, etc. ; 

 that the starches of different hybrids show a very wide 

 range in their parental relationships, some being almost 

 throughout very close to the seed parent, others very 

 close to the pollen parent, others for the most part inter- 

 mediate, etc.; that the starches of hybrids of reciprocal 

 crosses and of the same cross, respectively, are different, 

 the former differing from each other far more than the 

 latter from each other; that the relationships of the 

 properties of starches of hybrids to the properties of the 

 parents are in harmony with the data of the macroscopic 

 characters collected by Focke, with the data of DeVries 

 mutants (hybrids), and with the macroscopic and micro- 

 scopic tissue characters recorded in this research, in 

 showing that in any given hybrid the development of dif- 

 ferent characters m^ay take on different directions so that 

 some properties are like those of one or the other parent 

 or both parents, or developed in excess or deficit of 

 parental extremes, and also that new characters and 

 character-phases may appear. 



General Conclusions deawn feom Investiga- 

 tions OF THE MaCEOSCOPIC AND MiCEOSCOPIO 

 ChAEACTEES OF THE PlANT. 



The results of the studies of macroscopic and micro- 

 scopic tissue characters are in harmony with those re- 

 corded by Focke and of the researches with the starches 

 in showiug that in any given hybrid certain characters 

 may be the same as those in one or the other parent or 

 both parents, intermediate, or developed in excess or 

 deficit of parental extremes, and that the distribution 

 of these directions of character development is most vari- 

 able. A surprising result is found in a common lack of 

 correspondence between the percentages of macroscopic 

 and microscopic characters of any given hybrid that are 

 the same as those of the seed parent or pollen parent, 

 or intermediate, etc. Why, for instance, in any hybrid 

 the percentage of macroscopic characters that are the 

 same as those of the seed parent are relatively large in 

 comparison with the percentage of microscopic charac- 

 ters or vice versa is as yet inexplicable. What pertains 

 to one of the six parent-phases applies equally to all. 

 Moreover, there is not a constant quantitative agreement 

 between the macroscopic and microscopic characters, 

 separately or combined, and between either of these and 

 the starch characters of the same plant in the percentage 

 distributions among the parent-phases. 



The Relative Potentialities of the Seed 

 Paeent and the Pollen Paeent in Influenc- 

 ing THE ChAEACTERS OF THE HtBEID. 



The relative potentialities of the parents in determin- 

 ing the characters of the hybrids and in the distribution 

 of characters among the six parent-phases varies within 

 wide limits. In the starch reactions it is shown that in 

 some hybrids the influences of one parent are almost or 

 practically negligible, in others they appear to be about 

 equally divided, and in others there are various grada- 

 tions in degree and kind between these extremes. In the 

 tissue characters concordant results were recorded, but 

 here the variations were found to be very much restricted, 



