Stratigraphical and Faunistic Results. 245 
of the case, we have as a rule to speak of radioles and test-fragments by different 
names, still we have no right to assume that those found in the same rock repre- 
sent different species; therefore we ought not to base on the radioles any con- 
clusions more weighty than those supported by the test-fragments. 
If ever the geueral principles just laid down may be set aside, it is either 
when a sufficiently large number of radioles enables one to study the whole range 
of variation (as in the case of Anaulocidaris testudo), or when the material is well 
enough preserved to permit detailed examination and comparison of the micro- 
structure. The two methods ought, no doubt, to go hand in hand, but this is 
rarely possible with the Bakony material. Readers of the descriptions will, how- 
ever, have observed that every pains has been taken to apply these modern methods 
in arriving at the determinations. Fortunately also | have been able to study some 
of the rare specimens already described from other Raiblian localities, and to show 
a correspondence even closer than might have been inferred from the previously 
published names. 
It cannot too often be repeated that arguments based only on previous lists and 
records are rarely satisfactory. With the more minute subdivision of the strata, more 
precise description and determination of the fossils are required, and identifications 
that passed well enough at an earlier stage of the science have constantly to be 
subjected to fresh scrutiny. It is for this reason that it has been necessary to 
include in the present memoir so much discussion of species not found in Bakony 
at all, and for the matter of that not found in a good many other places where 
they have been said to exist. Jsocrinus propinquus and Anaulocidaris Buchi are 
notable instances. On the other hand I have refrained from the discussion of some 
records (e. g. PicuLer, 1857, 1866), when I have not examined the specimens on which 
they were based. This explains the patent omission of reference to isolated radioles 
of Cidaris alata, C. Roemeri, and C. cf. dorsata, recorded from the Raiblian of the 
Schlern plateau by Wourmann & KoKEN (1892, p. 171). The first and last of these 
may, for all one can tell, belong to the Raiblian mutations poculiformis and mar- 
ginata, while the supposed C. Roemeri might prove to be C. parastadifera. 
The names of species from the Trias of Kotel in Bulgaria, published in a 
preliminary note by Mr. P. BaxaLow (1905) sound interesting; but of course one 
can say nothing till the descriptions are published. His nomen nudum, Cidaris 
poculiformis, calls to mind the circumapical radioles of our Cidaris alata poculi- 
formis and C. dorsata marginata, and suggests that the age of the deposit may 
be Raiblian. His citation of Entrochus insignis Touta shows that these peculiar 
columnals, so reminiscent of the Cretaceous Pentacrinid Awstinocrinus are really 
Triassic, and not of that Cenomanian age to which they were assigned by TouLa 
(1890, p. 347, Taf. vi, ff. 3—6). Specimens from the type-locality, kindly sent me 
by Mr. BaxaLow (Brit. Mus. E 14076), suggest that the species is a passage-form 
between Eucrinus and either Jsocrinus or Millericrinus. There is nothing like it 
from Bakony. Of previously described Triassic forms that which most nearly 
approaches it is the much smaller Encrinus radiatus Scuaur. The other columnals 
figured by Toura (1890, Taf. vi, ff. 7—10) probably comprise the forms that BakaALow 
cites as Encrinus granulosus, Pentacrinus Fuchsti, and P. laevigatus. 
A special case is presented by the fauna of the Pachycardientuffe of the 
Seiser Alp. Its Echinoderm components were all referred by Dr. Brom to Cassian 
