November 3, 1922] 



SCIENCE 



517 



far from the attainment of this ideal, for, as 

 Parker says (p. 180), "The real difficulty lies 

 in the fact that the numerous receptors that we 

 now recognize have undergone varying degrees 

 of differentiation and hence their mutual affini- 

 ties are extremely diverse." 



On the morphological side the difficulties are 

 even greater, and the various attempts which 

 have heen made to determine which of the vari- 

 ous anatomical patterns of end-organs are more 

 primitive seem rather futile. Protoplasm in 

 general seems to he sensitive to all of the three 

 kinds of ^imuli of Parker's classification and 

 morphologically homologous organs seem capa- 

 ble of transgressing our artificial hdological 

 laws and "uniformities" in fashion most dis- 

 quieting to the systematist. This is illustrated 

 by Whitman's description (since confirmed by 

 Hachloy) of the cutaneous sensillae of leeches, 

 which are tactile in function on the body but 

 in the head become gradually metamorphosed 

 into visual organs, and also by the way in 

 which both olfactory and gustatory organs may 

 serve on occasion as either interoceptors or 

 exteroceptors, with characteristically different 

 central connections and reaction types in the 

 two eases. 



No better summary of this phase of the 

 matter can be given than the concluding sen- 

 tences of Parker's book: "It is because of the 

 repeated differentiations that characterize the 

 evolution not only of the chemoreeeptors but 

 of the other groups of like organs that a classi- 

 fication of them or even a simple enumeration 

 proves to be so unsatisfactory. For they are 

 not unitary elements that can be counted like 

 the fingers on the hand nor are they sviifieiently 

 co-ordinated to make classifications easy and 

 natural. They are like the whole organism 

 itself in that they exhibit that kind of diversity 

 that characterizes evolutionary flux." 



C. JuDSON Herrick 



SPECIAL ARTICLES 



PERIGENESIS 



I AM presenting here a short preliminary ac- 

 count of the results of a study of the division 

 figures in Tradescantia virginica L. 



With the methods used, the structure of the 



chromosome is that of an achromatic cylinder 

 of jelly-like consistency as described by 

 Vejdovsky (1912) in which the chromatin, 

 however, is imbedded in the form of chromo- 

 meres rather than a spiral. These bodies are 

 so distinct that in any one optical plane, they 

 can be counted. 



They are made up of flocculated chromatin 

 particles which associate together in rather 

 dense masses which are arranged inside of the 

 periphery of the linin cylinder in such a man- 

 ner that there results a central core of achro- 

 matic substance. 



The relationships of the chromomeres one to 

 the other seem to be somewhat variable 

 although the chromosome often shows a quad- 

 ripartite cross-section as figured by Merriman 

 (1904), Bonnevie (1908), and by Nawasehin 

 (1910). 



The effect of fixing, imbedding, and staining 

 this structure gives appearances which have 

 doubtless led to the in lerp retention that it is 

 longitudinally split. 



The arrangement of the chromatin particles 

 within the achromatic cylinder may be traced 

 back, in the vegetaitive stages especially, to the 

 earliest prophases and I do not find anywhere, 

 either in the vegetative or reduction divisions, 

 any further evidence of a longitudinal split so 

 that for the reductions, I agree with Meves and 

 others that there is no side by side pairing of 

 the chromosomes in these stages. 



I find as did Suessenguth (1921), in spite of 

 the recent evidence in favor of a parallel con- 

 jugation, that the continuous prophase spireme 

 is constricted into the chromosomes in end to 

 end relationships. 



Muller (1921), in discussing the work of 

 Troland (1917) says, "If he is right, each dif- 

 ferent portion of the gene structure must — 

 like a crystal — attract to itself from the proto- 

 plasm, materials of a similar kind thus mould- 

 ing next to the original gene another structure 

 of similar parts, identically arranged, which 

 then become bound together to form another 

 gene, a replica of the first." 



Prom the phenomena in all metaphase fig- 

 ures, inasmuch as I find the separation is not by 

 longitudinal division, I would limit the above 

 quoted process to the stages beginning with 



