July 23, 1920] 



SCIENCE 



83 



gation and independent assortment of factors 

 — has been shown to have a close parallelism 

 with the actual behavior of the chromosomes. 

 The situation is quite otherwise, however, as 

 to the mechanism of crossing over. Morgan 

 is right when he states that " while the 

 genetic evidence is favorable in all essentials 

 to the theory of interchange between homolo- 

 gous chromosomes, it must be confessed that 

 the cytological evidence is so far behind the 

 genetic evidence that it is not yet possible to 

 make a direct appeal to the specific mechan- 

 ism of crossing over on the basis of our 

 cytological knowledge of maturation stage." 

 Morgan, however, assumes the side-to-side 

 conjugation as a fact. His analysis of data 

 on parasynapsis leads him to the conclusion 

 that the early thin thread stage is most favor- 

 able for crossing over to take place. End-to- 

 end conjugation, or telosynapsis, according to 

 Morgan, " would have serious consequence 

 for genetics . . . , for while side-to-side union 

 offers an opportunity for interchange between 

 the paternal and maternal members of a pair, 

 no suoh interchange could be postulated if 

 end-to-end conjugation took place." 



It is the purpose of the present note to 

 emphasize that the process of end-to-end con- 

 jugation, at least as described by NothnageP 

 for a botanical object, and by myself^ for a 

 zoological one, does offer an opportunity for 

 crossing over to take place, contrary to 

 Morgan's statement. End-to-end conjugation 

 simply restricts, the stage in which such an 

 opportunity is offered. This can be readily 

 seen from the works of the above-mentioned 

 authors, who describe essentially the follow- 

 ing process: 



A separate loop or segment of double 

 spireme, whatever the nature of its duality 

 may be, gradually bends and halves of the 

 loop come to lie closely side by side. In the 

 tetrad thus formed there are four longitudinal 

 strands or threads. 



It will be seen, then, by telosynapsis, an 

 opportunity is offered for interchange between 



3 Notlinagel, M., ' ' Eeduotion Division in the 

 Pollen Mother Cells of Allium tricoccwm," Bot. 

 Gag., Vol. 61, 1916. 



chromosomes at the thick thread stage, but at 

 this stage only, in the manner originally sug- 

 gested by Janssen* in his chiasma type 



It must be remembered that the condition 

 of the chromatin threads at the early stage 

 when the double spireme develops is extremely 

 difficult to study minutely and accurately with 

 the method and apparatus at our command. 

 Under such circumstances, any inclination on 

 the part of the observer will have a consider- 

 able influence on the interpretation. If one 

 is so disposed, he may consider the condition 

 of the threads as representing the process of 

 pairing up. Dual threads develop out of 

 reticulum at this stage, and that was all I 

 could be sure of. There was certainly no 

 observable evidence of the process of pairing 

 up of two simple threads at least in the stone- 

 fly I studied. 



On the contrary, the formation of a tetrad 

 or ring by the bending of a loop of double 

 spireme, which appear in haploid number is 

 a clearly demonstrable fact. It is from this 

 ground that I interpret the haploid as being 

 composed of two homologous chromosomes 

 jointed up end-to-end, and its duality as in- 

 dicating primary splitting. No one has ever 

 seen two chromosomes actually coming into 

 conjugation, but the subsequent bending, re- 

 conjugation in side-to-side position, and the 

 ultimate segregation at metaphase, of the 

 halves of the loop is explicable only under the 

 assumption that two chromosomes were imited 

 end-to-end in the loop. 



Whether I am right in this interpretation 

 or not will be decided by future studies — per- 

 haps in very near future. Detailed compari- 

 son of the premeiotic stage with the prophase 

 of somatic mitosis would throw some light on 

 the situation. Also, a careful re-examination 

 of forms (Orthoptera, for instance), in which 

 parasynapsis is customarily claimed to occur, 

 with special reference to the haploid loops in 

 the thick thread stage would help settle the 

 question. Possibility no doubt exists that the 



* Janssen, F. A., ' ' La thSorie de la cihiasma- 

 tjpie. NonveUe interpretation des cinfeses de 

 maturation," La Cellule, T. 25, 1909. 

 theory. 



