434 
SUMMARY OF CURRENT RESEARCHES RELATING TO 
seen. Only in association with the sperm-nucleus was there a centro- 
some, which by dividing formed the two pole-corpuscles of the first 
segmentation-spindle. In detail, Boveri explains his grounds for 
believing that Fol’s “ quadrille ” was a misinterpretation. He believes 
still, as in 1887, that the centrosome of the Echinoid ovum is an organ 
which degenerates, and plays no part in development. He also discusses 
the observations of Guignard and Conklin, which support those of Fol. 
Centrosomes usually lie in the protoplasm, but they may, as Brauer 
has shown, lie within the nucleus. Similarly, the spindle fibres arise 
here from parts which lie in the nucleus, and there from parts which lie 
in the cytoplasm. Boveri holds fast to the conclusion that the centro- 
some is an independent and permanent organ of the cell. It divides 
into two, and forms the dynamic centre for the daughter-cells. 
The astrosphere is that complex of rays which is centered around the 
centrosome ; it is no permanent structure, but usually disappears com- 
pletely after division, arising quite anew in the preparation for the next 
division. But Boveri maintains, at the same time, his position that there 
is a specific substance — the archoplasm — composing the astrosphere. 
He proceeds to a stringent criticism of Heidenhain’s theory of cellular 
mechanics and his heresies in regard to the centrosome. One has hope 
that between two such masters a secure conclusion may eventually be 
reached. 
Ccelentera. 
Turbinaria.* — Mr. H. M. Bernard has some morphological and 
systematic notes on this very difficult genus of the Madreporaria. He 
urges that Milne-Edwards rejected the most natural character of the 
genus in abandoning the method of budding, and using in its place more 
artificial distinctions. A study of a series of the genus shows that the 
corallum of Turbinaria typically appears in its early stages as a small 
cup, but this cup-stage is generally transitory. It is at any rate clear 
that the cup-shape of the corallum must no longer be thought to be a 
specific distinction. A cross section through the stalk of a minute cup 
reveals a single, rather large, polyp-cavity surrounded by a thick spongy 
wall, which shows an irregular series of radiating plates bound together 
by irregular concentric synapticulse. This central polyp in the stalk is 
the parent polyp of the young corallum, and the spongy coenenchyma is 
a simple thickening of its walls. Surrounding the central cavity there 
is a series of longitudinal canals running parallel with the polyp-cavity. 
All these are in open communication with one another and with the 
polyp-cavity through pores. The axial polyp in the stalk of a minute 
Turbinarian colony buds laterally, and the buds form a simple ring 
round the axial polyp. The fundamental difference between Turbi- 
narians and Madreporarians is due to their different methods of budding. 
As soon as the polyjjs at any time forming the actual edge of the cup 
have, by radial growth, developed sufficiently from their next neighbours 
to admit of buds appearing between them, these appear, while the parent 
polyps bend sharply upwards towards the axis of the cup. As soon as 
this bending is effected, a fresh bud or fresh buds grow out close to the 
bend. It is thus by the continual addition of a fresh series of polyps 
* Ann. and Mag. Nat. Hist., xv. (1895) pp. 499-521 (2 pis.). 
