508 Mr. C. V. Burke on the Cyclogasteridae. 



resulted from tlie lack of an appreciation of the specific 

 characters and amount of specific variation. For instance, 

 we find that Cr/cloqaster liparis is described as having 

 D. 32-49; A. 26-38; P. 34-42. This amount of fin-ray 

 vaiiation certainly does not exist. It is doubtful if the 

 number of fin-rays in any of the species of Cyclogaster varies 

 more than six. It is probable that in Careproctus and Para- 

 liparis tiie number of fin-rays varies more than in Cyclogaster. 

 During the preparation of a monograph of the Cyclo- 

 gasteridas I have iiad occasion to attempt to untangle the 

 chaotic condition of the literature dealing with the species 

 represented in the museums of Europe. Tliis has been found 

 to be an impossible task without an examination of the 

 material. Such an attempt can lead only to greater con- 

 fusion. Tiie examination by a competent ichthyologist of 

 all the specimens of this family in the museums of Europe 

 would doubtless be very fruitful of results. I have found 

 eight undescribed species and a very aberrant genus in the 

 principal museums of the United States. The European 

 collections also doubtless contain undescribed species. From 

 the study of a large amount of material in the United States 

 it is possible to present a few notes which may aid any one 

 working with the specimens in the museums of Europe. It 

 is witli this hope that the present paper is published. 



A number of European writers have refused to recognize 

 the genus Careproctua, and have listed the species belonging 

 to this genus with those of Cyclogaster. The species of 

 Careproctus can be readily distinguished from those of 

 Cyclogaster by the absence of the posterior nostril. So far as 

 known, the two genera differ mainly in this single character. 

 Smitt (1893, p. 284) describes the adults of Cyclogaster 

 montagiii as lacking the posterior nostril, but this doubtless 

 is an error. Cyclogaster is the only described genus in which 

 both nostrils are present. In addition to the loss of the 

 posterior nostril, the species of Careproctus have obtained a 

 distinct coloration from those of Cyclogaster. Tlie coloration 

 may be considered a supplementary means by which we can 

 separate the species of the two genera. It probably will be 

 found to be of less importance than the difference in the 

 nostrils. The species of Careproctus are never variegated, 

 as is typical of the species of Cyclogaster. They lack the 

 mottlings, bars^ stripes, and blotches nearly always present 

 on the body or fins of the species of the latter genus. The 

 coloration of the species of Carejrroctus varies from whitish 

 to pinkish, dusky, and black. 



The species of Cyclogaster from the North Atlantic and 



