45 
considerally longer than those of the following myotomes. They 
extend ventrally and forward toward the branchial region, just posterior 
to which they unite into a common cell mass. 
In embryos of the 13th day, of 6 mm, the ventral longitudinal 
muscle of the gill region appears, in the position already described, 
as two deeply staining cellular strands, one on each side of the 
embryo. In most of the individuals I have studied, these extend from 
a point ventral to the third visceral ponch, where they end in a sharp 
point of the thickness of a single cell, backwärd through the branchial 
region, posterior to which they unite with the buds from the 
8th and 9th myotomes, which still retain connection with their 
myotomes. The cells of the bud from the 7th myotome have, however, 
lost connection with this myotome (as later do the ventral portions 
of the following myotomes as far as the 13th), and I infer that they 
have entered into the Anlage of the “Bauchmuskel”. In this stage 
the bud from the 10th myotome extends ventrally and forward to a 
point just behind that at which the buds of the 8th and 9th unite. 
Ultimately all the buds from the myotomes unite to form the con- 
tinuous M. par. ventralis, and I infer from its innervation in later 
stages that the buds from the 7th to 11th myotomes inclusive form 
that portion which primarily lies in the branchial region, to which 
for convenience I will apply the name M. par. subbranchialis. 
In a single individual of the 13th day of development the “Bauch- 
muskel” extends forward to a point beneath the 7th visceral pouch 
and its primitive relation with the post-branchial myotomes is clearer 
than in the individuals in which its anterior extent is greater. 
If we grant that the evidence stated above makes it probable that 
the M. par. subbranchialis has an origin similar to that of 
the corresponding muscle in Gnathostomata, how then are we to 
explain the incongruence in the segmentation of the dorsal and ventral 
musculature in the branchial region which has been commented upon 
by WIEDERSHEIM (80), JULIN (87), HATScHER (92) and Kuprrer (’95) ? 
In other Vertebrates a similar discrepancy appears (see MAURER, ’92) 
but, in Squalus at least, the number of segments in the hypoglossus 
musculature corresponds exactly with the number of myotomic buds 
from which it was derived. This is not the case in Petromyzon. 
HATSCHEK’s explanation, viz. “daß einige Myomeren an ihrem Hinunter- 
wachstum durch die Kiemenspalten behindert wurden und daher keine 
ventralen Teile lieferten”, has been found inadequate by KUPFFER 
because “thatsächlich wachsen die Myomeren in der ganzen Kiemen- 
region nicht weiter, als bis zu der lateralen Kiemenfurche, die die 
