ELONICHTHYS STRIOLATUS. 67 



PI. XV, fig. 1, represents a comparatively small specimen of the fish itself fronj 

 Burntisland, crushed upon its back, and wanting the caudal region. Fig. 2 of the same 

 plate represents the tail and hinder part of the body of a larger specimen from the same 

 locality, in which the extreme closeness of the articulations of the rays of the dorsal and 

 anal fins is well shown. Fig. 3 is a magnified representation of one of the scales of the 

 same example. 



B. Variation intennedius. — As already mentioned, I gave in 1877 the name 

 Eloniclithys bitertnedius to two of the three specimens which Agassiz erroneously in- 

 cluded under Amblypterus punciatus, namely, to those figured in j)l. 4 c, figs. 3 and 5, of 

 the atlas to vol. ii, pt. 2, of the * Poissons Eossiles.' These two specimens were 

 imperfect, the head and anterior part of the body being wanting in both ; but the same 

 locality soon supplied a good many more or less entire fishes which were clearly identi- 

 fiable with them. As I have devoted the whole of PI. IX and part of PI. X to these 

 Wardie s[)ecimens, I may, before going further, indicate the salient points which they exhibit. 



In PI. IX, fig. 1, we have a specimen in a Wardie nodule, which shows the general 

 form very well, being quite entire with the exception of the caudal fin, which is slightly 

 injured; but this deficiency is supplied by fig. 2, which shows the entire caudal in addi- 

 tion to the dorsal and anal, and is, in fact, a fresh draAviiig from the specimen figured by 

 Agassiz, op. cit., pi. 4(7, fig. 3. It will be seen that the form of the body and the pro- 

 portions of the fins are the same as in striolatus, but the articulations of the longer rays 

 of the dorsal, anal, and ventral fins are further apart, the joints appearing consequently 

 longer than broad, as shown magnified in fig. 8. These articvdations are, however, pretty 

 short in the j)Cctoral fin, as represented in PI. X, fig. 4, slightly magnified. Figs. 3, 4, 

 and 5, Pi. IX, represent the external aspect of scales from the other of Agassiz's originals 

 {op. et tab. cit., fig. 5), fig, 3 being an ordinary flank scale, fig. 4 one from the lateral 

 line on the flank, and fig. 5 one from the middle of the caudal region above the anal fin, 

 and it will be seen that their sculpture is essentially similar to that on the scales of 

 striolatus. But in most of the Wardie specimens which occur in nodules the ganoine layer 

 becomes separated from the scale, and adherent to the counterpart when the nodule is 

 opened, disclosing the subjacent osseous tissue, which is marked by more or less concen- 

 tric coarse ridges and furrows, as seen in figs. 7 and 8 — an appearance which must not be 

 confounded with the true external sculpture. In PI. X, fig. 2, we have some of the bones 

 of the right side of the head shown in impression. The operculum (op.) is rather narrow^ 

 its posterior-superior angle rounded off", while the posterior-inferior one is acutely 

 pointed; the suboperculum {s. op.) is square-shaped. The exact number of the branchio- 

 stegal rays {dr.) is hardly ascertainable ; only a few are shown here, but in other speci- 

 mens I he anterior one of each lateral series is distinctly shown to be nuich broader than 

 the rest, and a median lozenge-shaped plate is also conspicuously present between these 

 and behind the symphysis of the jaw. In the specimen here figured the pre-opcrculum, 

 the cheek-plates, and tlie maxilla are gone, whereby the upper end of the hyomandibular 



