Dr. Gerhard Holm — 0)% the Structure of Graptolites. 439 



accepted opinion concerning the presence or absence of a virgula 

 in all Graptolites, although such was never described or expi'essly 

 mentioned, except in the groups Diplograptidaj, Monograptidfe, and 

 Retiolitidge, might be explained by the two following circumstances: — 

 One of these is that Barrande only knew of forms with a virgula, 

 with the exception of Retiolites, in which genus he certainly had 

 observed the chitinous filament named by later authors "a straight 

 virgula," but referred it to the outer network. He began his chapter 

 on the virgula (solid axis) thus : " Graptolites are always provided 

 with a virgula."^ Through his authority he might so have influenced 

 later authors that his representations concerning the virgula became 

 a creed. The other circumstance might be that broken branches of 

 Dichograptidge have long been confused with species of Monograptus 

 in which the virgula is really present, hence the absence of the 

 virgula in the former has never been questioned. 



How strongly the idea of the presence of a virgula in all true 

 Graptolites has taken root is best shown by the fact that so keen 

 and critical an observer as Brogger^ was disposed to admit the 

 presence of a virgula in Dictyonema and Bryograptus, although he 

 had never observed it in these forms. 



Genus DiDYMOGRAPTus, M'Coy. PI. XIV, Figs. 1-3, 7, 8. 



Concerning the structure of the proximal part of Didymograptus, 

 one finds hardly anything mentioned in literature, and in the figures 

 of its species the proximal part is in the greater number of cases so 

 inconspicuously or indefinitely indicated, and figured only from one 

 side, that no definite conclusions can be drawn as to its structure. 

 One exception is Moberg's description of Didymograptus gibberidus, 

 Nich. — "Nyagrapt. fran Skane," p. 339, pi. viii — for which, 

 however, Moberg, in ignorance of the true structure of the Didymo- 

 graptus polypary, and misinterpreting the characters in the 

 incomplete material at his disposal, founded a new genus, 

 Isograptus. As will be seen below after the description 

 of the Didymograptus polypary, Moberg's description and figures, 

 properly interpreted, show clearly that the proximal part of 

 Isograptus gibberulus completely agrees with that of the genus 

 Didymograptus. Some observations about the Didymograptus 

 polypary are mentioned further by Tornquist (Siljansomradets 

 Grapt., i, p. 15). Concerning the sicula, Tornquist remarks that it 

 seems sometimes as if it " had been divided first by an oblique wall 

 into two parts, and as if each part sent out a separate branch." 

 The figures of the proximal part — figs. 8, 12, and 14 — which all 

 show the sicula side, indicate that it is the sicula and left theca 

 joined, together with the walls between them, which are referred to 

 by Tornquist. Tornquist states further that the sicula " keeps its 

 triangular form, and by a conspicuous wall is separated from both 

 stipes," which always arise " at different levels from the sides 

 of the sicula, and seem to have originated by a kind of budding." 



1 Grapt. de Bolieme (1850), p. 4. 



2 Die Silur. Etagen, 2 u 3, 18S2, p. 37. 



