100 PHYSICAL BASIS OF HEEEDITY 



to place themselves at this time in pairs. The double 

 threads shorten later to take on the form of the ordinary- 

 chromosome. How the earlier, long thin thread (lepto- 

 tene thread) is changed into a thick thread when the 

 chromosomes condense is not known. According to sev- 

 eral accounts the thread coils spirally within the wall of 

 the ' ' chromosome, ' ' at first in a loose coil, then in a tightly 

 twisted coil. This idea of a coiled thread, or core, in a 

 condensed chromosome is one that fits in very well with 

 the idea that the thin thread represents the essential ele- 

 ment in the chromosome that retains its original continuity 

 even when the chromosome is condensed into a short rod 

 or even into a ball. Unfortunately the evidence in favor 

 of this view is by no means well established. 



At the time when the threads conjugate, the evidence 

 in several forms, such as Batracoceps, Tomopteris, etc., 

 shows that when the conjugating pairs are U-shaped, the 

 union begins apparently at both ends of the U at the same 

 time. When the chromosomes are rod-shaped (in the last 

 telophase) the evidence fails to show whether the union 

 begins at both ends simultaneously or at one end only. 



As the union between the threads progresses the parts 

 not yet united can often be seen to be twisted about each 

 other. They not only overlap, but they seem to be wrapped 

 around each other. 



Whether the threads are split lengthwise before their 

 union can not be stated for all cases. It is certain that 

 no splitting has been seen in several animals, but in one 

 case (Ascaris) the threads have been found to be split 

 lengthwise before they conjugate. 



For a short time following the union of the threads 

 they come in close contact with each other, and give the 

 impression of having fused into a single thread. Usually 

 before the nuclear wall breaks down to release the thick 

 threads, a split can be seen again extending throughout 

 the length of the threads. Not infrequently another 

 longitudinal split appears in each half resulting from the 



