Herrick, Werves of: Siluroid Fishes. 225 
Attis in his earlier paper (’89) incidentally referred to the 
lateral canals of Ameiurus catus and states that the opercular 
canal does not communicate with the main canal. WRIGHT 
(84, p. 264) says that in A. catus there is an occipital com- 
missure just cephalad of the post-temporal bone by which the 
main canals of the two sides communicate and that this com- 
missural canal bears two pores near the middle line. Mc- 
Morricu (’84 a, p. 271), likewise describing A. catus, says of 
the supra-occipital bone, ‘‘Posteriorly on each side of the 
fontanelle, it presents many minute foramena, belonging to the 
system of mucous canals.”’ The inference from these descrip- 
tions is that in A. catus there is a supra-temporal canal running 
in the occipital bone, a condition which I do not find in any of 
the types examined by me. Whricut in the passage cited 
asserts that the opercular canal does not communicate with the 
main canal. CoLiinGcE, describing the lateral canals of 
Ameiurus catus, affirms ('95, p. 279) that there is such an 
occipital commissure across the mid-dorsal line (without, how- 
ever, giving any description of it), and also that there is a tree 
communication between the opercular and main canals, and he 
figures a series of ‘four small drainpipe-like canal bones. which 
pass from the region of the posterior border of the hyoman- 
dibular bone to the lateral border of the frontal (Pl. XVIII, 
fig. 3, c. 6.) Passing from the main canal into this series of 
canal-bones, and through the external portion of the hyoman- 
dibular bone, the canal enters the preoperculum, from here it 
passes into the distal portion of the quadrate and then into the 
mandible, opening by four pores in its course.’’ There are 
several anomalies in this description to which we shall have to 
return later. I may now merely mention in passing that in 
this figure the bone marked sub-operculum is of course the 
inter-operculum the sub-operculum of siluroids, as is well known, 
being represented in the first branchiostegal ray. Finally ALLIs 
(97, p. 632) acknowledges his error in the matter of the connec- 
tion between the opercular canal and the main canal in Ameiurus 
and adds, ‘‘As I find the canal in Silurus.glanis as he gives it 
in Amiurus, he is undoubtedly correct.’’ 
