OF THE SALP^. 39 



formerly connected is absent. Thus the Solace may be 

 divided into two classes, the associated and the solitary. 



With regard to these strange creatm-es, Chamisso* 

 made the ciu-ious observation, that the solitary Salpce, 

 though never themselves forming parts of a chain, yet 

 ahvays contain a progeny resembhng those which do, and 

 that each separate embryo is linked together, like the in- 

 dividuals forming the Salpa chains, and very closely re- 

 sembles them in figure; but in the associated Sal^a on the 

 contrary, he found young, resembling the solitary ones. 

 This was usually a single pedunculated foetus, situate in, 

 and attached to, the wall of the respiratory sac of the 

 parent animal, and resembling the solitary SalpcB in its 

 shape. In only one associated Salpa {ßalpa zonaria) 

 there were always found four such pedunculated distinct 

 embryos. Now, as examination of the free swimming- 

 individuals, which presented marks of having been dis- 

 jointed from a chain, proved that these also contained 

 only sohtary pedunculated embryos, Chamisso concluded 

 from his observations, that all solitary Saljpce, produced 

 associated ones or chains ; and on the contrary, that all the 

 associated Salpce were parents of solitary ones, and these 

 again of the associated and so on. 



The generations of the SaljJce consequently, are alter- 

 nately solitary and associated, so that a Salpa mother, to 

 use Chamisso's familiar expression, is not like its daughter 

 or its own mother, but resembles its sister, its grand- 

 daughter, and its grandmother. Against this theory of 

 the alternate generations of the Salpa, its intelligent 

 author had many attacks to sustain, and the more so, on 

 account of the considerable difference which exists be- 

 tween individuals of the associated and solitary Salpce, 

 and because no example of such alternate generation was 

 known in the rest of nature. Thus it was no wonder, 

 that endeavours were made to subvert this theory, partly 

 by seeking to weaken the force of the observations upon 



* Chamisso. Dc animalibus quibusdam e classe verniium LiiiiiEeanä. Fase. I. 

 de Salpa. Berolini, 1819, 4to. 



