10 



G. HOLM AND A. H. WESTEEGAAED 



the cephalon the new subspecies agrees almost completely with the latter, there being only 

 a small difference in the lateral outline, which in A. pisiformis is posteriorly curved inwards, 

 in A. pisiformis pater almost straight and parallel to the axis. A small but clear distinction 

 is to be seen in the pygidium, the axis of which in A. pisiformis is almost devoid of trans- 

 verse furrows, but in A. pisiformis pater is divided by distinct furrows into two rings and 

 a large end-lobe, which is faintly swollen but not to the same extent as in A. pisiformis 

 ohesus Belt. Marginal spines of the pygidium are very short; they are not visible in the 

 Bennett Island specimen, probably owing to imperfect preservation. 



A. pisiformis pater is met with in the Pa/radoorides forclihammeri zone of Scania and 

 the Agnostus laevigatus zone (?) of Nerke. Infrequent. It is no doubt older than A. pisi- 

 formis proper, the first specimens of which appear in the zone of A. laevigatus} 



Agnostus glandiformis Angelin 



[PI. I, fig. 2-5; pi. IV, fig. 1-3] 

 1851 Agnostus glandiformis, Angelin, Palaeontol. Suecica, p. 5, pi. 6, fig. 1 (pygidium only). 



1878 

 1880 

 1901 

 1902 



Beogger, Paradoxidesskifrene ved Krekling, p. 58, pi. 6, fig. 7. 



TuLLBEEG, Agnostusarterna vid Andrarum, p. 29, pi. 2, fig. 20. 



LiNpsTKOM, Researches on the visual organs of the trilobites, p. 37, pi. 1, fig. 7. 



(pars), Gbonwall, Bornholms Paradoxideslag, p. 63 (not the pygidium figured, pi. 1, 



fig. 6). 



A couple of cephala and pygidia are present, which may be identified with Agnostus 

 glandiformis. The resemblance seems to be complete,'^ even as regards size. One of the 

 pygidia (fig. 4a — c) shows part of the doublure, which in this species is peculiarly 

 developed, as is to be more clearly seen in the Scanian specimens reproduced for the sake 

 of comparison on pi. IV, fig. 1 — 3. 



If the cephalon figured by Angelin as A. glandiformis is correctly depicted, it does not 

 belong here, since it has a distinct marginal rim, which according to Angelin's description 

 and as shown by complete rolled up specimens, is absent in this species. Tullberg (1. c.) 

 has suggested that Angelin has confused the cephalon of A. glandiformis with that of 

 A. hitubermlatus when depicting the said species, as the figures do not agree with the 

 diagnoses. It is true that Angelin's diagnosis of the cephalon of A. glandiformis agrees better 

 with his figure of the cephalon of A. hituberculatus and vice versa, but only to some extent, 

 for the basal lobes which characterize the latter according to diagnosis and figure, are 

 lacking (or at least very indistinctly developed) in the former species. Therefore, the dis- 

 agreement between diagnosis and figure may be ascribed to the latter's not having been 

 carefully drawn. 



1 Following a manuscript by J. C. Moberg, C. Wiman 

 (Sver. geol. unders., Ser. A^a, No. 5, p. 92) has quoted 

 Agnostus pisiformis from a layer of dark stinkstone at 

 the boundary between the P. tessini beds and the P. 

 forchhammeri zone of Southern Oland. According to my 

 own researches in the matter of the sequence at the loca- 

 lities in question, this stinkstone layer may belong to 

 the conglomerate of BUlingseUa exporrecta. The latter, 

 which in Southern Oland replaces the zones of (P. da- 



vidis?), P. forclihammeri, Agnostus laevigatus, and the 

 lowest part of the Agnostus pisiformis zone, is at places 

 developed as a stinkstone without a distinct conglomeratic 

 structure but crowded with fossils, of which Agnostus 

 pisiformis is by far the most abundant. 



2 Whether the small puuctiform tubercle of the py- 

 gidium, which is always to be seen in well-preserved 

 Scandinavian specimens, is lacking or developed in the 

 Siberian form, cannot be stated. 



