Dr. 0. JB'ernnann — Distribution of GraptoUtes. 451 



the true Graptolites or Graptolite-like organisms will furnisli a 

 proof of their former free mode of life. 



In many genera {Didymographts, Dicellograptus, Coenograptiis, etc.) 

 the sicula runs out into a point as sharp as a hair ; in other forms 

 (e.g. Dichographts octobracMatus. Hall, Galgenberg. Christiania) it has 

 acquired the form of a round knot or knob ; in other species again 

 (e.g. species of Tetragraptus) it entirely disappears in full-grown 

 individuals, although the spot at which we ought to look for it 

 is quite uninjured. All this speaks against a former attachment. 



We now assume that all Graptolites provided loith a sicida were not 

 attached bodies. 



It is indeed conceivable that we have an analogy with existing 

 Coelenterata, in which also attached forms (these would correspond 

 to the arborescent Graptolitiform organisms) are known, as well as 

 forms which are inserted into the mud, and perfectly free. ^ But 

 whether they pierced into the mud of the sea-bottom with the sicula, 

 or whether the Graptolites kept at small depths like a Cartesian 

 diver in a vessel of water, will depend upon how far down in the 

 deep seas of that time, the element of the Graptolites, organic 

 existence was possible. 



Angle of Divergence. — Before we advance to the definition of 

 the angle of divergence in the branched Graptolites, we must make 

 out clearly what part of the hydrosoma we are to regard as the sicula 

 in the different groups of Graptolites, seeing that the sicula and 

 angle of divergence stand in the closest relation to each other. 



In most genera, as, for example, Didymograptus, M'Coy, Dicrano- 

 graptns, Hall, Diplograptus, M'Coy, and Monograptus, Gein., there 

 is no doubt about this, but not so in the genus Dicellograptus, Hopk. 

 The situation of the sicula here has been a question in dispute, and 

 indeed because in that genus there are often four processes at the 

 point where the two branches grow together, one of them situated 

 on the dorsal side of the branches, and three upon the cell-bearing 

 ventral side (Fig. 3). Hopkinson, who in 1871 ' separated the genus 

 Dicellograptus from the Didymogra'pti, considers the middle one of 

 the three spines which are placed side by side to be the sicula 

 (" radicle "), and defended his idea warmly against Nicholson, who,^ 

 in some of the forms now referred to Dicellograptus {e.g. D. anceps, 

 Nich.), regarded the process on the dorsal side of the branches 

 (Hppkinson's " axillary spine") as the embryonal piece. Nicholson, 

 himself, however, is not consistent in his mode of designation, 

 seeing that in other evidently nearly allied forms (e.g. D. divaricatus, 

 Hall) he regards the middle one of the three spines, in agreement 

 with Hopkinson, as the sicula. Carruthers, in opposition to 

 Hopkinson, regards the spine in the axillary hollow of the branches 

 as the sicula ; and Lapworth also expresses himself in the same 

 sense, writing as follows : ^ — " It is simply this persistent sicula 

 which constitutes the ' axillary spine ' in Dicellograptus." 



' "On Dicellograptus, a new gemis of Graptolites," Geol. Mag. Vol. VIII. 

 (1871), No. 1. 2 » Monograph of the British Graptoliticte " (1872), p. 55. 



3 "Notes on British GraptoUtes," Geol. Mag. Vol. X. (L873), p. 501. 



