14 



G. HOLM AITD A. H. WESTERGAAED 



owing to the peculiar shape of the axis of the pygidium. As the Swedish specimens are far f 

 better preserved than are those from Bennett Island the following description is founded on 

 the former.^ 



Cephalon fairly arched; marginal rim of moderate width, produced in a short spine at 

 the genal angle. Glabella totally lacking a transverse furrow, provided with a slight keel; 

 basal lobes of moderate size, triangular. Cheeks nearly equal in width throughout, confluent 

 in front of the glabella. 



Pygidium subrectangular; margiilal rim slightly extended behind the axis, for the rest 

 narrower and of uniform width, produced posteriorly into two small spines. Axis rather 

 wide, continuing to the rim, without transverse furrows, anteriorly convex and raised into 

 an elongated tubercle, posteriorly deeply depressed and having a pair of longitudinal 

 impressions. 



A small form; the largest cephalon met with is 2.5 mm in length and width, the 

 largest pygidium 2.75 mm long and 2.5 mm wide. 



The specimens from Bennett Island referred to this species vary, as is seen from the 

 figures. The pygidia, so far as they are preserved, agree fairly well with the Swedish form, 

 as is also the case with the cephala in fig. 35 — 37. The latter was regarded as a pygidium 

 by Holm, but in my opinion it may be the cephalon of an immature individual of A. repandus 

 at a stage before the basal lobes had developed (Barrande's second or third stage). A couple 

 of cephala, especially those of fig. 42 and 43, difi'er from the typical form in being wider 

 and having a narrower glabella. These dissimilarities may be explained in part by different 

 preservation: the Swedish specimens are preserved in limestone and show the original form 

 of the test, the Siberian ones are preserved in shale and are more or less flattened. But 

 this alone is not enough entirely to explain the differences, as, for instance, the cephalon 

 in fig. 43 is inconsiderably more flattened than that in fig. 35. However, the wide form 

 does not appear to be strictly distinguished from the narrow one, as intermediate forms 

 seem to, exist. Consequently, I do not think it proper at present to separate the wide form 

 as a distinct species or subspecies. 



The systematic position of A. repandus is doubtfiil, and it seems to present affinities 

 to different groups of Agnostidae. As regards the cephalon, it resembles the groups A. parvi- 

 frons and A. fallax to some extent, though it is sharply distinguished from the former by 

 its far longer glabella and from the latter by its lacking a transverse glabellar furrow. 

 In its long and non-lobate glabella (apart from the basal lobes) it is of a singular type. The 

 pygidium especially resembles an und escribed species from the zone of Agnostus laevigatus 

 of Vestergotland, which, in so far as the pygidium is concerned, is just intermediate between 

 A. incertus Brogger and A. repandus. 



A. repandus appears in the Paradoxides forchhammeri zone (Andrarum limestone) at 

 Andrarum in Scania, but it is extremely rare there. A few specimens are met with in 

 a boulder of stinkstone at Skyllersta, province of Nerke, belonging to the Agnostus laevi- 

 gatus zone (?). 



1 There can be no doubt that the cephalon and the 

 pygidium belong to the same species since a couple of 

 cephala and pygidia of the kind in question have been 



found on a bedding plane of a stiukstone boulder from 

 Nerke, yielding no other Agnosti than A. laevigatus, 

 A. pisiformis pater and A. planicauda. 



