Corresjjondeoice — L. F. SpatJi. 143 



fossil forms suggests a more secure attachment than in the living 

 Nautilus) some additional means of attachment was useful, and that 

 this was suj)plied by (but not the only function of) the penetrating 

 fibres of the folded margin in Ammonoids and the sijohuncle in Nauti- 

 lus. Since the animals with severed siphuncle mentioned by Willey 

 were attached to their body-chambers by means of the shell-muscles 

 (Willey performed the operation at the base of the body-chamber, 

 behind the annuhis) there was, of course, no danger of the animals 

 falling away from the shells. Truncation of the septate portion is 

 not imcommon in fossil forms, and in some Discoceras (probably 

 on adaptation to a purely crawling mode of life) all the. camerae 

 were cast off and the shell of the adult (and not till then proj)agating) 

 animal consisting of the body-chamber only. This merely proves 

 that there was no further formation of septa, and the shell-muscles 

 became permanently attached to the shell-Avall and held the animal 

 fast, but it is not evidence against my view. 



On the other hand, it could be argued that owing to its con- 

 striction at the septal necks, the siphuncle of even a truncated 

 Protohactrites or of an AiupJioreojJsis (with only a few camerae) 

 might have been sufficient for the jDurpose of attachment. Only here 

 it must be admitted that if this notching of the siphuncular tube 

 is lookedupon as a strengthening feature for such attaching purposes, 

 then a similar function must also be assumed for Ammonoidea. In 

 some large sections before me {Parhinsonia dorsetensis, Phylloceras 

 heterophylluni) this constriction at the septal neck is more apparent 

 than in the (dried) shells of the recent Nautilus. On the other hand, 

 inflation of the endosiphotube (between successive endosipho- 

 sheaths) and of the ectosiphuncle (between the septa) is found in 

 many fossil Nautili (I am not referring to the actiniform siphuncular 

 structures), and the separation of these from the camerse by either 

 thick mineral deposits or by continuous septal necks does not 

 favour an assumption of gas-secretion by the vascular siphuncle. 



My opinion certainly was not based on the interesting, if un- 

 scientific, account quoted by E. A. Smith. A reference to this 

 was given merely to show how the shell-muscles in the living 

 Nautilus became detached. But the origin of the siphuncle as a 

 constriction of the viscera, and the various structures in the 

 multitudinous developments of the order Nautiloidea, especially 

 the more primitive forms (Piloceras and the Proteroforms of the 

 Cameroceras-Vaginoceras series) seemed to me most instructive, 

 as also the transitions from the endosiphuncular structures to the 

 later ectosiphuncle of e.g. Baltoceras, which genus Hyatt placed as 

 the first of his family Orthoceratidae. I collected a number of very 

 interesting Lower Ordovician Cephalopoda in Newfoundland some 

 seven years ago, but their description is delayed because it entails 

 a revision of the classification of the whole order Nautiloidea. I may 

 also mention that the blackness of the outer coat of the siphuncle 

 in Piloceras (the specimens are preserved in a dolomite, exactly like 



