200 Annals (jf the Carnegie Museum. 



The maxillary is contained 2.33 in the head, while in the females it 

 is almost 3. Of two males a couple of inches smaller, one is inter- 

 mediate in these characters, and the other does not at all differ from 

 the females. These are possibly characters appearing with age after 

 the limit of size is reached. 



54. Xiphistes chirus (Jordan and Gilbert). 



This species was not at all common about the islands; Xiphidion 

 miicosum being much commoner, which in turn was not nearly so 

 common as Xiphidion riipcstre. 



While collecting in 1895 in the southern end of the Sound and at 

 Cape Flattery this species was found abundantly at the former 

 locality, and one specimen (the type of Xiphistes ulvce) was taken at 

 the Cape, while the other two were taken onl}^ about the Cape. All 

 three of the species were taken the following year at Port Ludlow, but 

 the relative abundance of each does not appear. Further information 

 on the distribution of these fishes in the sound would be of interest. 

 It is not a deeper water species (at least about the islands) than the 

 others as alleged in the original and in current descriptions, being 

 found wherever Xiphidion nuicosuvi occurs. 



A re-examination of the type of Xiphistes ulvce proves almost beyond 

 doubt that the species is untenable. The branches of the upper, lateral 

 line are not longer than may be found in individuals of Xiphistes 

 chirus. It has, however, three anal spines, but as all other characters 

 are identical with those of Xiphistes chirus it is doubtless abnormal 

 in this character. Color, the only other alleged difference, is so vari- 

 able in these fishes that little dependence can be placed upon it. 



55. Lumpenus anguillaris (Pallas). 



Many specimens were seined from shallow water. The young have 

 elliptical spots arranged in longitudinal rows, about equal in color, 

 size, and spacing, over the sides and back. In the adult the spots in 

 the row along the lateral line become more elongated and distinct, 

 while below them the spots disappear, and above them they are more 

 or less broken up into clouded reticulations. 



The statement in current descriptions of the genus Lumpenus 

 that the lateral line is indistinct, or obsolete, does not adequately 

 state the facts, at least for this species. The lateral line is represented 

 by slightly enlarged scales, not -inconspicuous in individuals of mod- 



