12 G. HOLM AND A. H. WESTEBGAAKD 



Andrarum limestone, for instance, indicates that at least this species is very variable 

 as to length and width of cephalon as well as of pygidium. The pygidia from Bennett Island 

 too, with some hesitation referred to the Scandinavian species, agree with it fairly well even 

 if the marginal rim in the Siberian form is less conspicuously extended at the sides than it 

 is in the latter. 



As cited above the pygidium and probably also the cephalon of A. hitubercidatus were 

 erroneously regarded by Gronwall as young specimens of A. glandiformis: 



The Norwegian form, which was identified as A. hituherculatiis by Brogger, is no 

 doubt a distinct species, as pointed out already by Tullberg and Gronwall. Professor Qron- 

 wall has had the kindness to lend me for examination some specimens from Krekltng, which 

 were labelled by Brogger as A. hitvherculatus Ang. and presented to the Geological 

 Institution of Lund in 1878. These specimens agree very well with the figures given by 

 Brogger and are readily distinguished from A. hituberculatus, as described and depicted by 

 Angelin, inter alia by theii' having axial furrows in cephalon and pygidium and lacking 

 distinct basal lobes of the cephalon. The Norwegian form in question is here called Agnostm 

 confusus n. nom. (pi. IV, fig. 7, 8). It is met with in the Paradoxides forcJihammeri zone 

 and also in Scania, where however it is very rare. 



A. hituberculatus is known from the zone of Paradoxides forchhammeri in Scania 

 (Andrarum, Kiviks-Esperdd), Angermanland and Bornholm. It is infrequent in Scandinavia. 



Agnosfus nudus Jiyperloreus n. subsp. 



[PI. I, fig. 6-9] 



Some specimens have been tentatively brought together under the above name. As 

 indicated by the figures they show some differences, but the variability does not seem to be 

 greater than it is, for instance, in A. nudus ma/rginatns Beogger. 



The resemblance between the form in question and A. nudus (Beyeich), as described 

 and figured by Barrande, may justify us in regarding it as a subspecies of the latter. It is 

 distinguished from the Bohemian form in having the marginal rim of the pygidium somewhat 

 narrower and of almost uniform width. In A. nudus the rim, according to Barrande, is always 

 wider behind than at the sides of the pygidium, a characteristic still more accentuated in 

 the related Scandinavian form A. nudus marginatus. 



The cephalon of A, nudus hyperhoreus may easily be confused with that of the associated 

 species A. glandiformis. However, there seems to exist a slight difference in the outline, 

 which in the former is somewhat more subrectangular owing to its fairly truncate front. 



A. nudus occurs in Barrande's etage C of Bohemia and in the Menevian of Wales. 

 A. nudus marginatus belongs to the upper part of the Paradoxides davidis zone (and 

 the P. forcJihammeri zone?) of Scandinavia. 



Agnostus latirhachis n. sp. 



[PL I, fig. 14—23 (24, 25?)] 



The pygidium is characterized by its wide, imsegmented, slightly conical, bluntly 

 terminating axis, which is provided with a faint elongated tubercle; lateral lobes confluent 

 behind the axis; marginal rim of moderate width, not dentated. ' 



