SIPHUNCLE AND PALLIAL VESSELS. 759 



aber nur mit einer. Somit ist das Innere vom Sipho als die directe Fortsetzung des 

 grossen Cblomraumes aufzufassen, und das Colomepithel setzt sich als ein niedrig- 

 cubisches, beinahe plattes Epithel in die Siphonalhbhlung continuirlich fort." 



The relative importance of a correct knowledge of the nature of the siphuncular 

 cavity is too obvious to render an apology for the quotations which I have given 

 necessary. In fact this cavity is a venous cavity or haemocoel, as I have satisfied 

 myself by means of transverse and longitudinal sections, but above all by dissection, 

 which besides showing the non-existence of any opening into the coelom has revealed 

 the presence of a hitherto undescribed venous channel, which I will call the posterior 

 pallial vein, into which the siphuncular vein, i.e. the axial cavity of the siphuncle, 

 discharges (PI. LXXVL, figs. 7 and 8). It is quite probable that the posterior pallial 

 vein which runs above, approximated to, and parallel with, the posterior pallial artery 

 in the genito-intestinal ligament, is identical with the "free passage" which Owen 

 described as coursing alongside of the posterior pallial artery between the pericardium 

 and the siphuncle. As a matter of fact the channel which occurs in this position is 

 a vein which runs forwards until it arrives at the pericardium, when it bends up above 

 the dorsal wall of the latter and enters the central venous sinus from whence the 

 afferent branchial vessels are given off. 



The finer structure of the siphuncle (PI. LXXXIII.) is complicated and not readily 

 intelligible. Surrounding the central vein is a loose meshwork of trabeculae, the meshes 

 communicating with one another and with the central vein. The communications 

 between the peripheral system of spaces and the central sinus are best seen in longi- 

 tudinal sections, but are also to be found in transverse sections. The external epithelium 

 of the true or pallial siphuncle which secretes the partially calcified and partially horny 

 siphuncle, which remains behind after the soft parts are extracted, presents a very 

 peculiar structure. This has been in part correctly described by Haller, who states 

 {op. cit., p. 201) that the entire surface of the siphuncle is thrown into very fine 

 longitudinal folds due to the plaited membrana basilaris of the epithelium, and in each 

 plication of the basement-membrane there is a narrow space lined by flattened endo- 

 thelium (Haller, op. cit, Taf. XII., fig. 19). 



I can confirm the existence of these folds, although they do not appear in my 

 preparations to be so regular as Dr Haller describes and figures them. Moreover I cannot 

 see the regular deep epidermal pits which alternate in correspondence with the subjacent 

 mesodermal folds in his figures. What seems very remarkable is the fact that the 

 spaces in the folds are in reality prolongations from the meshes of the trabecular tissue 

 of the siphuncle, and therefore represent intra-epidermal blood-spaces. I think this 

 circumstance must throw some light on the function of the siphuncle, which is essentially 

 a vascular appendix. I have already expressed the possibility that the siphuncle would 

 be capable of keeping the air in the chambers up to the normal pressure by making 

 good losses caused by diffusion. The same, or a closely similar suggestion, was in fact 

 made by Keferstein (1865, op. cit. p. 374)'. 



Coupling the existence of intra-epidermal venous spaces with my observation recorded 



1 Cf. Meigen, W., " Ueber den hydrostatischen Apparat des Nautilus pompilius." Arch. Naturg. 36 Jhg. 

 Bd. i. 1870, pp. 1 — 34. This author confirms Keferstein's theory. 



