27 



The median limbs of the loops, converging at the apical plate, do 

 not have the long, transversely directed continuations as represented 

 by Morgan's fig. 10 (vd. fig. I a). They terminate in a manner more 

 nearly like those of the New England Toimaria , shown by the same 

 author in fig. 7. The second point of diff'erence that I mention is the 

 small loop in the transverse post-oral portion of the longitudinal band 

 on the side of the larva 90° from the mouth, fig. 1 /'. In the narroAv 

 neck of this loop there appears to be the anläge of a single pair of 

 processes like those found in the preoral portions of the band. So far 

 as I am aware this loop does not exist in any other Tornaria. 



The posterior circular ciliated band, so characteristic of Tornaria, 

 is present and bears cilia much longer and stouter than those of the 

 longitudinal bands. And in addition to this there is a second circum- 

 anal band about mid way between the large one and the an vis. This 

 second band is quite inconspicuous, and, so far as I have been able to 

 determine, is without cilia. In section the cells of the ectoderm in 

 this second band are found to diff'er from those of other regions some- 

 Avhat as do those of the large one. I am inclined to think that this is 

 a rudimentary band. In the possession of a second circular band 

 this Tornaria agrees with the one described by Metschnikoff 1870. 

 (The larva represented in fig. 1 is so situated with reference to the 

 observer, that neither of the circular ciliated bands is visible.) 



The eye spots are well developed, and are farther apart than those 

 of the Bahama species, figured by Morgan. The following measure- 

 ments were made on a larva nearly jjeady to undergo metamorphosis. 

 The specimen had been preserved in chromic-osmic solution : Length 

 1,9 mm; greatest thickness 1,33 mm. This is very nearly if not quite, 

 the maximum size. It is, therefore, considerably smaller than the Ba- 

 hama form, this latter being 4 mm long, according to Morgan. 



This larva presents two points in its internal structure that are 

 particularly worthy of note. The oesophageal evaginations which are 

 the beginnings of the first pair of gills, do not appear until the meta- 

 morphosis sets in. In the oldest Balanoglossus that I succeeded in 

 getting, but two pairs have yet appeared, and neither of these has yet 

 fused with the ectoderm, preparatory to breaking through to commu- 

 nicate with the outside world. Of these two pairs the first is much the 

 further advanced, the formation of the tongue-bars having begun. In 

 this respect it agrees essentially with several other species, but difi'ers 

 strikingly from the New England form in which, as is well known, 

 four pairs of gills have begun to develop by the time the metamor- 

 phosis begins. 



