58 ONYCHIOPSIS. 



Sphenopteris Manlelli type," and adds: "It resembles more than 

 any other described fossil Thyrsopteris elongata, Geyler." If we 

 look at Fontaine's pi. xxvi. figs. 6 and 7, and compare with 

 Geyler's pi. xxxi. fig. 4, it is well-nigh impossible to agree that 

 these are different species, to say nothing of distinct genera ; or, 

 again, it is difficult to separate, on good grounds, fig. 2, pi. xliv. 

 of Fontaine from Onychiopsis elongata, as figured by Yokoyama, 

 pi. ii. fig 2. It is true that Fontaine's specimens show in some 

 cases a stouter rachis than occurs in the Japanese examples. 



There seems to be but slender evidence in support of 

 T. ITeektana, var. angustiloba. Cf . 0. elongata ; e.g. Fontaine, 

 pi. xliii. fig. 8, and Yokoyama, pi. ii. fig. 2. 



The species of Onychiopsis described by Yelenovsky, 1 referred 

 to below, should be compared with T. crenata, pi. xxxix. fig. 1, 

 Fontaine, and Yelenovsky, pi. i. fig. 9. 



In T. densifolia, Font., we have a fern which seems to be very 

 similar to Yokoyama's fig. 2, pi. ii. of 0. elongata ; the difference 

 consists in the somewhat broader pinnules of the former. The 

 figures of T. densifolia suggest a form of plant identical with that 

 represented by T. Meekiana, Font., e.g. pi. xxxviii. fig. 3. 



Another of these "Thyrsopteris" species, T. microphylla, is 

 referred by Fontaine to the " Sphenopteris Mantelli type," and 

 the figures of this species appear to me illustrations of the plant 

 included under the name T. angustifolia ; both come near to 0. 

 elongata. Compare also T. pinnatifida, Font., pi. Ii. fig. 2, etc., 

 with 0. elongata, and with the latter species T. incequipinnata, 

 Font. 



T. elliptica, Font,, is described as "not very close to any 

 previously described fossil ; " 2 there appears, indeed, to be a rather 

 close resemblance between the figures of this species, pi. xlvi. 

 fig. 1, and T. demifolia, Font., pi. xl. fig. 4. In addition to the 

 two species T. Viryinica and T. alata, to which allusion has been 

 made as examples of specific determinations which are without 

 such claim as entitles them to serious recognition as well-defined 

 types, a third may be mentioned, T. heterophijlla, Font., which does 

 not appeal more strongly for acceptance, if we may judge by the 

 figured fragments, pi. Iviii. figs. 3 and 3a. 



1 Loc. cit. p. 10, pi. i. figs. 6-12. 



2 Fontaine, loc. cit. p. 133. 



