390 
comes very large and the heart is commonly drawn out to a filamentous 
form. This continues to beat until the death of the embryo, but does not 
handle any blood. The eyes do not attain their full size, and may be poorly 
pigmented. They often are abnormally set so that they occupy the fore- 
part of the head. This may fuse into a median single eye or may be pres- 
ent on one side only. The ear vesicles often become large and inflated, giv- 
ing rise to a large rounded prominence on each side. The pigment cells 
are very finely developed, show a tendency to a pattern and bilateral sym- 
metry, but there is a lack of uniformity in this in the different embryos. 
The embryos are shortened and may develop abnormally large pectoral 
fins. It,is not necessary to give more than a general description at this 
place. 
Even within this group it is very easy to distinguish between the 
hybrids in which the Menidia notata is used as the male from those in 
which the Menidia gracilis is the male. The development of the former is 
more successful in those that pass the blastopore closure stage, although 
my experiments show that the mortality is greater at this point. The pig- 
mentation is better developed and the various organs above mentioned 
are laid down much more normally. As a consequence fewer and less pro- 
nounced abnormalities occur. In the Fundulus-Menidia gracilis cross it is 
not uncommon to have only one eye formed. This may be lateral or may 
be median. The eyes are commonly set much further anterior so as to 
occupy the front of the head than in the nearly related cross. 
In the crosses of group 3 we obtain quite a different series of 
hybrid embryos. None of these will develop as far along as those in 
group 2. There is the usual large mortality preceding and at the blas- 
topore closure stage. The more successful embryos are much shorter, the 
pigmentation is much less perfect, the black usually predominates, the eyes 
are never normal, and often altogether wanting, and the life of the embryo 
is shorter. The heart and pericardial cavity is much the same as in the 
Menidia hybrids, although I have seen no attempt to develop vessels on the 
yolk. Their embryos show in every way that the developmental processes 
have deteriorated much earlier than in the Menidia crosses. 
When we come to group 4 we have a still more prenounced abortion 
of the developmental processes. Many of the embryos close the blastopore 
after a fashion, but the embryo is always much shortened, usually being a 
mere streak of protoplasm. These embryos do not lengthen to form a 
