EARLY STAGES OF DEVELOPMENT OP BALANOGLOSSUS. 153 



mud, has less need of bands of cilia and organs of sense than a 

 larva which leads a free existence at the surface. 



Less intelligible are the absence of a water- vessel and of a 

 contractile string. With regard to the former, Spengel (7) states 

 that in Tornaria it eventually forms the bodj^ cavity of the 

 "prseoral lobe and its pore persists as the proboscis pore. It 

 becomes, therefore, certain that the antei'ior body cavity of the 

 opaque larva, which has a similar fate, is homologous with the 

 water-vessel of Tornaria. Moreover, in Tornaria the water- 

 vessel opens to the left of the middle line, while in this larva 

 the external opening when it appears is connected with the left 

 posterior horn of the cavity. The origin of the middle and 

 posterior pair of mesoblastic sacs is similar in the two larvae, 

 and their eventual disposition in the opaque larva corresponds 

 generally with the fate described by Spengel for the two pairs 

 of archenteric diverticula in Tornaria. 



Since MetschnikofF (6) has recently published a paper in which 

 he proposes to class Balanoglossus together with the Echin- 

 odermata in a common class " Ambulacralia," and since this 

 proceeding is mainly supported by arguments adduced from a 

 comparison of Tornaria with Echinoderm larvae in general, and 

 with a typical Asteroid larva in particular, it will be necessary 

 to see what new light the existence of a second type of Balano- 

 glossus larvae throws on this thesis. Until the later develop- 

 ment of the opaque larva is known it is impossible to make any 

 such comparison, but it is perhaps worth pointing out that while 

 this larva differs from Tornaria in many points it happens that 

 all of these points (if we except the presence of eyespots) are 

 those in which Tornaria resembles an Asteroid larva. Never- 

 theless, most of these features, the praeoral and longitudinal 

 postoral bauds of cilia, &c., are very possibly purely secondary 

 structures whose presence is correlated with a pelagic habit of 

 life. If this is true, however, they could not avail either to 

 connect or to separate Tornaria from the Asteroid larvae, and 

 the existence of an opaque Balanoglossus larva which lacks 

 them would have no significance in settling this question. 



Tlie absence of any proper water vessel and contractile 



