Correspondence. 403 
Endoceras beUmnitiforme Holm presents the same generic characters 
an JVun/to </i/fiiiai C\-dvkc. In the belemnite-like specimens (Holm, loc. 
cit., pi. I, figs. 5a-5c, and Clarke, loc. cit., pi. VI, figs. 5-7) the apical 
cone represents, according to Holm, the initial chamber of the Endoce- 
ras; the contracted surface, the first septum; and the elongated portion, 
the siphuncle. The thickened wall or "apical solid cone," as described 
by Clarke, is noted also by Holm in E. belemnitiforme (loc. cit., p. 8. pi. 
I, fig. 2); and that the sfpho is also in contact with the outer shell, and 
even fastened to the same, is both described and figured. 
Very interesting is the interpretation made by Holm of this species of 
Endoceras, the only species of which the initial chamber was known, 
and in the structure of which he finds the suggestion that the initial 
chamber and the sipho of Endoceras species with large siphos served ai 
first, and to a great extent continually, as a visceral chamber. From 
the characters presented by this species, Holm thinks that the Endoce- 
ras type is the primitive fossil cephalopod, and that the Tetrabranchi- 
ata were yet earlier derived from forms having an open conical shell 
without septa or sipho. The striking similarity between E. belemniti- 
forme and Nannoaulema permits exactly the same interpretation from 
either. 
The "siphonal cone," as composed of siphonal sheaths described by 
Clarke, is, on the other hand, not observed by Holm; but, according to 
Holm's interpretation, it might be expected that the shell at the apex 
would be continually thickened while the initial chamber and sipho 
were still occupied by the viscera. Tin- existence of such a structure 
seems to support the theory given by Holm,* that the primitive Tetra 
branchiata filled the entire conical shell; that afterward, following ex- 
tended growth of the shell and elongation of the animal, an air chamber 
enclosed by a septum was formed upon one side of the animal: and thai 
by the formation of a succession of similar septa- originated the elon- 
gated sipho. This, which was occupied by part of I he viscera at h'rsi. 
as in some species of Endoceras, has, in the cephalopods with small si- 
phons, been reduced to a narrow, perhaps functionless structure. 
The name Nanno, which is seemingly an inappropriate one for a ge- 
nus of Cephalopoda, is, according to other aul hors. onlv a synonym for 
Endoceras; but the new species described is of great interest. 
Frederick \Y. Sardeson. 
Geohgisch-mineralogiscJies Institut der Uhiversitat, F^reibwrg i. limit n. 
Oct .'II, IS'.'J,. 
Evidence of Sdperglacial Eskers in Illinois and northward. 
The paper by Mr. .1. I!. Woodworth on the origin of eskers, noticed on 
page 396, and the work of Mr. (i. II. Barton on the same question, also 
there cited, with the discussion of superglacial drill bj Prof. It. I). 
Salisbury in the last number of the Journal of Geology (vol. n, pp. 613- 
<>:;■>. Sept. -Oct., 1894), make this an opportune time for directing atten- 
*See also Steinmann and Doderlein, "Elemente dei Palaontologie" (1890), p. 
