374 
peculiar structures (Coenogametes) are 
likely to be found far more generally in 
this group of fungi than has been imagined, 
and the subject takes on considerable bio- 
logical interest. 
The fusion of multinucleate gametes re- 
sults in the pairing of the sexual nuclei and 
consequent union, two by two, giving about 
half as many fusion products as the original 
number of sexual nuclei. The phenomenon 
recalls the conjugation of swarm spores in 
water outside of the parent gametangium, 
the difference being that here the sexual 
elements are retained in the parent struc- 
ture. 
The origin of these coenogametes is un- 
certain. It was suggested that they may 
have come from a much simpler condition 
of sexual organs than has been supposed. 
They are possibly derived from a gametan- 
gium of the lowly type presented in several 
algal groups where motile gametes are dis- 
charged into water. An alternative hy- 
pothesis demands that the behavior is excep- 
tional and derived from a condition of 
heterogamy.. But a serious objection will 
be raised to such a view if the conogametes, 
as seems likely, are found to be far from 
exceptional among the Phycomycetes. 
Early Development in certain Hybrid Species . 
W. J. Mornxkwaus. 
In more than 20 crosses among fishes, 
there was not a single failure of impregna- 
tion, although some of them were between 
species of different orders. The per cent. 
of impregnation ranged from 1 to 95. The 
highest per cent. may be between the most 
distantly related species. Eggs normally im- 
pregnated complete the segmentation stages. 
Two crosses and their reciprocals largely 
went to closure of the blastopore with the 
main axis of the embryo apparently nor- 
mally formed. Only the closely related spe- 
cies went far enough to hatch. The cross 
between brook trout ? and lake trout re- 
SCIENCE. 
(N.S. Vou. XIII. No. 323. 
sulted in ‘fingerlings,’ a large per cent. of 
which had the caudal peduncle aborted 
and many had the anal fin wanting. 
The rate of development was in all cases 
that of the egg species or slower. The con- 
jugation of the pronuclei is normal. In the 
first cleavage spindle of the cross between 
Fundulus heteroclitus 9 and Menidia notata 
the @ chromosomes appear as long, slightly 
wavy rods, and the ¢ chromosomes as short, 
comma-shaped rods. This difference is 
maintained to the 16-celled stage, as far as 
they have been followed. In the cross be- 
tween Fundulus heteroclitus 9 and Ctenolab- 
rus adspersus the é chromosomes are scat- 
tered along the spindles. This condition 
obtains as far as the third cleavage and there 
is some evidence that some or all of the ¢ 
chromatin may be lost in the course of 
development. 
Abnormalities on the Horny Plates of the Turtle, 
Chrysemys marginata: 8. E. Merk. (Will 
appear in the American Naturalist. ) 
Variation in the Madreporic Body and Stone 
Canal of Asterias vulgaris: GERTRUDE C. 
DAVENPORT. 
A number of cases of multiple madre- 
poric plates were observed in the starfish 
Asterias vulgaris, of Cold Spring Harbor. 
This multiplicity may occur either in one 
interradius, in several of them, or in all. 
Thus three plates were observed in one and 
the same interradius and in another case 
five. In one case five plates were observed 
in one interradius, and one in another of 
the same starfish. A six-rayed individual 
had one madreporic plate in each of the 
six interradii. When only one madreporic 
plate is present it may be lobed or partially 
divided by cross seams into 2, 3, 4, 5 and 6 
parts. 
The multiplicity of plates may or may 
not be accompanied by a multiplicity of 
stone canals. A group of four plates had 
only one canal, while one of five had two. 
