109 



The development of the gonodendra could be followed step by step. The very first 

 indication of these appendages is to be seen at the base of the young siphons which are 

 situated on the first turn to the left which the stem performs in its first proximal part 

 underneath the pneumatophore. They are just visible with the naked eye as small white 

 spots. Another is drawn lower down on PI. XIX; it has already increased in size. The gradual 

 increase of the gonodendra is visible on the base of the intermediate siphons; they were 

 omitted in the drawing PI. XIX. 



Going down A another gonodendron with a microscopic gonostyle is seen near the 

 base of a lower siphon. 



Going up B one sees clearly near the base of every siphon how the gonodendron 

 increases in size side by side with its gonostyle. In A tentacles are not yet developed. These 

 show themselves for the first time at the base of the 6''^ and following siphons as small 

 curved tubular appendages. 



The same gradual development as in the gonostyles can also be traced in the tentacles. 

 Whilst in B they are small stumps, we see how in the distal part of C they attain a leno-th 

 of more than one cm. Tentilla were never observed. The shape of the tentacles and their 

 position at the base of the siphon entirely- agree wath the same in Pterophysa grandis which 

 we had the opportunity of examining and which had been identified by Fewkes himself. 



B. B has a length of 38 cm. The siphons increase more and more in size, the utmost 

 length being 37.. cm. They are all transparent, more or less curved, the ptera on each side 

 have the same development as in A. The tentacles show themselves for the first time as 

 is said above, the gonodendra have attained a dimension of i mm. in diameter, one or two 

 of the most distal ones possess a small gonostyle. 



C. C attains a length of 30 cm. The stem is somewhat narrower, but does not show 

 any difference in structure from A and B. 



The siphons undergo a considerable alteration. They become less transparent, more 

 muscular especially near the aperture; near the base the transparency vanishes entirely, 

 probably through the greater development of the entoderm. On the whole the siphons 

 get a more regular "siphon"-aspect, and only now probably their normal function of 

 assimilating food, which they receive through their mouth-opening, commences. As the 

 transparency of the siphons diminishes so does also the breadth of the ptera. (Compare 

 PI. XXIV, figg. 167, 168, 169). On PI. XIX this is not clearly indicated as a space of 

 15,5 cm. is left out between C and D but still on comparing the siphon ".v 2" of C with 

 siphon ".f i" of B one is struck with the extreme difference in shape. In D, E, F, G and 

 the intermediate parts, which were not drawn, no trace of ptera is left. Therefore we arrive 

 at the following conclusion : 



The siphons, which are quite as transparent as their ptera, 

 which latter show a considerable development (at least -/.; of the whole 

 breadth) and possessing moreover an apical aperture which has been 

 found by making microscopic sections, should be considered as the 

 locomotive appendages oi Pterophysa. Through expansion and contraction 



