310 PROCEEDINOS OF THE NATIONAL MUSEUM. vol.39. 



members of this middle class are not as yet well enough known to be 

 located with precision. Much more work must be done upon the 

 ecology, morphology, and especially upon the ontogeny of the various 

 families, before any lasting suggestions can be made in the system- 

 atization. 



The present paper is intended as a contribution toward this end, 

 and if it succeeds in placing the family Ergasilidse upon a solid basis 

 of morphological and ontogenetic facts it will have accomplished its 

 purpose. With reference to the systematization the author has at 

 present only three suggestions: 



1. The Lichomolgidse should not be included in the same family 

 with the Ergasilidse, as is done by Thorell and Gerstaecker, but the 

 two should constitute separate families, as given by Sars and T. 

 Scott. The family Ergasilidse as thus constituted includes only those 

 forms which are parasitic on fish. 



2. The Ergasilidse are closely related to the Corycseidss and Licho- 

 molgidse on the one side and to the Chondracanthidse on the other, 

 the four families forming a series from a free-swimming condition 

 through the various stages of commensalism, semiparasitism, and 

 parasitism to the degeneration and modification which always result 

 from extreme parasitism. These four families ought therefore to be 

 kept close together in the classification scheme. 



3. The Ergasilinse, the Bomolochinse, and the Tseniacanthinse are 

 so closely related and form such a natural series that they should be 

 grouped together rather than separated into distinct families. The 

 nature of this relationship appears clearly in the following discussion. 



FAMILY DIAGNOSIS. 



Body with cyclops form, segmentation perfect or nearly so, first 

 segment fused with head, other segments free or fused among them- 

 selves, genital segment enlarged but little, abdomen narrow, anal 

 laminae with very long setse. First antennae tactile, never prehensile 

 or locomotor, and alike in both sexes; second pair strongly developed, 

 prehensile, and armed with stout claws. Mouth-parts degenerate, 

 better suited for piercing or tearing, but with no trace of a siphon; 

 maxillary hooks present in some species and like those in the Caliginae. 

 First four pairs of legs biramose, rami distinctly jointed; fifth pair 

 uniramose, with indistinct joints. 



Sex organs paired and situated in the cephalothorax ; egg-sacks 

 like those of Cyclops, eggs multiseriate and numerous. 



Females becoming more or less fixed parasites, males remaining 

 free-swimmers tliroughout life. 



