14 ARTHUR BOLLES LEE 
telophase. This next stage will be most conveniently studied 
in the spermatogonia and oogonia of the Amphibia. For 
here, as the clump passes into the telophase, it expands into 
a wide ring, on the surface of which the chromosomes are set 
on widely spaced meridians, figs. 48, 44, 45, 48, 49, 50, and 
others. Owing to this arrangement they show only a minimal 
amount of overlapping, and, standing out on a clear back- 
sround, can be studied with sufficient accuracy. 
In the earliest stages of this process of expansion (figs. 35 to 
38) we find much the same state of things as in the denser clump. 
The paired chromosome axes can be more clearly distinguished : 
periaxial spirals can be just detected on some of them, and 
on others their existence is placed beyond all reasonable doubt 
by the lateral processes visible on the edges of the axes. And 
the sheath can be made out on many of them (same figs.). In 
later stages such as figs. 39 to 47, the demonstration of these 
details becomes more difficult, mainly on account of two com- 
plications which here ensue. One of these is the formation 
of trabeculae (‘ anastomoses’ of some authors) between the 
chromosomes. These trabeculae obscure the lateral pro- 
cesses, with which they are easily confused, and so deprive us 
of an important guide for the detection of the periaxial spirals. 
The other is, that as the clump expands, the chromosomes 
elongate ; and as they elongate their duplicate axes t wine 
round one another, figs. 85, 89 to 47.1 This involves 
1 This gives us the key to Kowalski’s assertion (op. cit.) that the chro- 
mosomes of the salamander larva are at certain periods alveolated. Thirteen 
of his figures purport to show the alveolesin question. Eight of these are 
of telophases. On comparing them with my figs. 39 to 51 it becomes 
evident at once that Kowalski has interpreted images of doubled and 
entwined chromosome axes as borders of alveoles—which is very natural, 
for a thus doubled chromosome easily gives the impression of an alveolated 
cylinder if you are not able to obtain a sufficiently sharp focusing of its 
entwined axes. The remaining five of Kowalski’s figures of ‘alveolated’ 
chromosomes are of spiremes, such as my figs 25 to 27, and manifestly only 
show that the chromosomes he had before him were double, transverse 
trabeculae uniting their two moieties being taken for transverse walls of 
axial cavities in an undivided cylinder or riband, 
