GYEODAOTYLID^. ' 49 



" parent " in respect of size ; wMlst, moreover, it contained within 

 its interior another very young Gyrodactyle, or, in other words, a 

 *' grand- daughter." Professor Van Beneden, as we shall presently 

 see, interpreted these facts very differently, but, it must also be 

 noted that Yon Siebold's ideas had gained a very general accept- 

 ance. For my own part, I incline to the views of Yan Beneden ; 

 but provisionally accepting Yon Siebold's conclusions — so far, at 

 least, as G. elegans is concerned — I may proceed to remark that I 

 also have noticed the second generation, or daughter, to contain in 

 its interior evidences of what I, at the time, supposed to be a third 

 generation.* This phenomenon I observed in specimens of the 

 worm obtained from the tails of Gasterostei caught m the Serpen- 

 tine, Regent's Park. Indications of the third progeny were seen 

 whilst the daughter still resided within the body of the nurse-parent, 

 but the so-called grand- daughter acquired much greater conspicuity 

 immediately after birth. In one instance, I had a very satisfactory 

 opportunity of watching the process of separation. The " daughter" 

 commenced showing itself externally by a slight bulging at the centre 

 of the body of the parent, whilst the integument of the latter yielded 

 on all sides of the bud-like projection, and in such a manner as to 

 convey the idea of a vaginal opening, although a sheath of this kind 

 was by no means actually visible. There was an evident struggle 

 on the part of the young one to free itself from the so-called 

 parental envelope, but the tissues exhibited no evidence of injury. 

 On partial protrusion, it was seen that the budding portion corre- 

 sponded with the centre of the daughter's body, and this, in a little 

 while, assumed the aspect of a semi- circular, or horseshoe- shaped, 

 band. Subsequently the upper or anterior end became detached, 

 the freed extremity being now recognized as the head, characterized 

 by the possession of two ventricose lobes. A considerable interval 

 elapsed before the broad posterior end of the animal could be dis- 

 engaged, but immediately after this was effected the sides of the 



* As stated in my paper on Gyrodactylus in the " Microscopical Jonmal," new series, 

 vol. ii., p. 36, 1862. --T. S. 0. ' 



H 



