Material. Of the two examples of this form, specimen A is complete the petals and apical disc being covered with secondary calcite, and the details of the base obscured; specimen B is broken (much of the adapical surface is missing), but shows the plating-structure of the base. Where not covered with a layer of secondary calcite, the specimens are weathered and the ornamentation partially destroyed; secondary calcite, deposited irregularly, often obscures the scrobicules and pores. For comparison, I have examined four dried specimens of the modern form, A. recens, in the Zoological Department of the British Museum (Natural History); my thanks are due to the keeper of that Department, and to Mr. C. C. Monro, for permission and facilities to investigate these specimens.

Locality. North end of Red Bluff, six miles north of Waitangi.

Description. The dimensions in mm. are:—

Test depressed, adapical surface and ambitus rounded. Ambital outline oval to subpentagonal; the line of greatest width is excentric posteriorly. Periproct oval, longitudinally elongated; at the anterior end of the sulcus beginning at the ambitus (but not notching it). Peristome invaginate, subpentagonal, excentric anteriorly. Adoral surface undulate; interambulaera 1 and 4 are raised into slight prominences near the ambital margin. Ambulacrum 3 is in a broad shallow depression; the two similar depressions in which are situated ambulacra 1 and 5 coalesce near the peri-stome, by the suppression of the ridge into which interambulacrum 5 is raised towards the ambital margin. The size and shape of the petals are shown in figure 1; the outer pore of each pore-pair is slightly elongated and larger than the inner. There appear to be 24 and 25 pore-pairs respectively in each column of ambulacrum 3, and 26 to 28 in each column of ambulacra 1 and 5, and about 23 in each column of ambulacra 2 and 4. So much as can be deciphered of the plating-structure is shown in text-figures 1–7. In the apical disc, the posterior oculars appear to meet; irregularities in alternation are seen in the adapical parts of petals 2 and 3 (text-figure 2). The plates within the peristomial invagination are obscured by secondary calcite; the erratic weathering makes it somewhat difficult to determine the pores, which are usually filled with secondary calcite. The adapical transverse sutures of the plates bordering the peristome are to some extent hypothetical as shown in text-figures 3–7, but are based on the position of the pores; the adradial sutures are, however, quite distinct. The first three plates in each column of the paired ambulacra bear three or four pores in a straight line; these form the adoral portions of what are here termed the “inner columns” of pores. Above this, the plates are of three types:—

Plates of types A and B bear pores near the adradial suture, forming the “outer columns” of pores; plates of type C near the peristome bear their pores near the centre, and their pores thus continue the “inner columns” Traced towards the ambitus, the “inner columns” slope adradially to join the “outer columns”; when the columns merge, the pores are arranged triserially. Ambitally and adapically (outside the petals), the pores are irregularly uniserial. In ambulacrum 3 there is no distinction into “inner” and “outer columns”; on the adoral surface, the pores are triserial, but are uniserial and central near the peristome. There are no bourrelets.

The tubercles are weathered; but the better preserved are perforate, and some retain traces of crenulation.

Comparison with A. recens. The specimens closely resemble the modern New Zealand species Apatopygus recens (Milne Edwards). The values (relative to the length) of the width and height of the test, and of the dimensions of the periproct and peristome, fall within the range of variation exhibited by the specimen figured by Agassiz (op cit.) and by the examples in the British Musuem (Natural History); except that specimen B is slightly narrower. The peristome and apical disc are very slightly more excentric anteriorly than in the modern form. The shape of the adoral surface agrees with that of the British Museum specimens. The apical disc of A. recens has been figured by Mortensen (1922, op. cit., pp. 187–8 text figures 21 a, b), and by Agassiz (1873 op. cit., pl. 14a, figure 2), and shows a certain amount of variation. The apical disc of specimen A is of the same type as that figured by Mortensen in his text-figure 21a, in the protrusion of the oculars 2, 3 and 4 into the corona, and the complete separation of oculars 2 and 4 by the madreporite; but in Mortensen's figure the posterior oculars are separated, the genital pores extend outside the apical disc, and the madreporic pores encroach on the left anterior genital plate. In the figure given by Agassiz, the posterior oculars meet, as they appear to do in specimen A.

The ambulacral plating structure is identical in type with that described by Hawkins (1920, op. cit., pl. 7, figs. 1, 2) in A. recens, although there are differences, which examined in the light of Hawkin's morphogenetic theory (discussed below), possess varying degrees of significance. Thus the “trifling imperfection of alternation” in petal 3 of the modern form which “may well be but an individual irregularity” (Hawkins, op. cit., p. 396), is seen in both petals 2 and 3 of the fossil (text-figure 2). Of possibly more significance, is the absence of “Discoidiid” plating adorally in

ambulacra 1 and 5 in specimen B. Differences in the distribution of “Pyrinid” plating are much more significant. In ambulacrum 4 of the fossil form, “Pyrinid” plating does not reach the ambitus (text-figure 6); unhappily in the Recent form examined by Hawkins the plating - structure of ambulacrum 4 was obscure in the region of the ambitus, although “Pyrinid” plating “seems not to reach the petals” (Hawkins, op. cit., p. 396). In ambulacrum 3 of the Recent form, “Pyrinid” plating reaches the petals; in the fossil form it certainly reaches nearly to the ambitus (text-figure 5); adapically the structure can not be ascertained, but there are indications that demiplates are absent. In ambulacra 1 and 5 of specimen B, of the 30 plates in each column below the ambitus, there are from 8 to 10 “Cidaroid” plates above the “Pyrinid” plates (text-figures 3, 7); in the recent form (Hawkins, op. cit., pl. 7, figure 1) in ambulacrum 1, there are below the ambitus 5 and 3 “Cidaroid” plates in each column respectively, out of 38 and 40. Thus in ambulacra 1 and 5 “Pyrinid” plating affects the fossil form to a less extent than the living form; and it is at least possible that a similar relation obtains in the other ambulacra.

The arrangement of the pores round the peristome has been described by Mortensen (1922, op. cit., pp. 184–5), and described and figured by Hawkins (op. cit., p. 395 et seq., pl. 7); the disposition as seen in two specimens in the British Museum (Natural History) is illustrated in text figures 8–12. In both these examples, the arrangement of the pores in ambulacra 2 and 4 may be regarded as intermediate in type between that in ambulacra 1 and 5, and that in ambulacrum 3; futher, in ambulacrum 3 the triserial arrangement of the pores appears at the fifth pore in each column, counting from the peristome. In all other respects, the fossil and the recent form here figured are similar. Mortensen states (op. cit.) that there are four columns of pores in all the “phyllodes”^* in the specimen of the recent form which he examined; so that in this respect the recent form has in some examples the arrangement seen in the fossil form. Moreover, in the figure given by Hawkins^† of ambulacrum 3 in the Recent form (Hawkins, op. cit., pl. 7, figure 2), the triserial arrangement of the pores appears at the seventh pore, as in the fossil (text-figure 5). It would seem, therefore, that the arrangement of the pores in the fossil can be paralleled amongst the variation exhibited in the living form.

The arrangement of the tubercles is similar in both the living and the fossil form. Mortensen states (op. cit., p. 188) that the tubercles in A. recens are perforate and crenulate; the tubercles in

the British Museum specimens, and in the fossil form, are also perforate and crenulate.

Of the various differences between the fossil and the living form which have been discussed above, the most important is probably the difference in the extent to which “Pyrinid” plating affects ambulacra 1 and 5. That the ambulacra of the fossil form are less affected than those of the living form, affords support to the hypothesis, suggested by Hawkins (op. cit.), that A. recens is morphologically the end member of a series in which “Pyrinid” plating progressively affects the ambulacra to an increasing extent. Since, however, our knowledge of the plating-structure is founded on but one example of each form, and we are ignorant of the amount of variation which may exist in each, it seems better not to create a new species for the fossil form.

Comparison with other species. The genus Apatopygus is at present known to contain but one species. Of Australasian forms which superficially resemble A. recens, Nucleolites papillosus Zittel^* is distinguished easily by the posterior sulcus, which nearly reaches the apical disc; the plating-structure (and consequently the true generic position) is unknown.

Discussion of Apatopygus and its affinities. The arrangement of the pore-pairs adorally in A. recens as described by Hawkins, and by Mortensen, and as here figured, are all to some extent different. In the specimens I have examined, the pores are arranged in four columns in ambulacra 1 and 5 (text-figures 8, 12); in ambulacra 2 and 4 text-figures 9, 11) the pores are triserial, but with a distinct tendency towards the arrangement in four columns, which is described in A. recens by Mortensen, and here figured in the fossil form (text-figures 4, 6).

It may be suggested that, in the specimen figured by Hawkins (op. cit., pl. 7, figure 1), the pores of plates 10 and 12 in column 1b, and of plates 5 and 12 in column 1a (counting from the peristomial margin) may be regarded as pores of the “inner columns”; the irregularity of the arrangement of the pores (as compared with that in specimens examined by Mortensen and myself) may be due to “irregularities” in the plating-structure—the interruption of the succession of “Pyrinid” plates by two groups of “Discoidiid” plates in column 1a, and the consequent “irregularity” in the adjacent parts of column 1b. Whether the normal arrangement in the posterior ambulacra of A. recens, is a regular succession of “Pyrinid” plates with four columns of pores, or the less regular arrangement with “Discoidiid” plates (as described by Hawkins), can only be determined by the examination of more material.

In either case, the diagnosis of Apatopygus, and the description of A. recens, must be revised in the light of the descriptions given here and by Mortensen, and of the examination of the fossil form.

Hawkins, in discussing the affinities of his genus Apatopygus (op. cit., Geol. Mag., 1920, pp. 397–8), recognises in the Nucleolitidae (s. str.) four types which “illustrate a morphogenetic sequence”:—

Comparison of the figures quoted above shows that in this morphological sequence, in the hypophyllodes^* of Nucleolites and the corresponding areas in the other genera.

(a) the arrangement of the pores is modified. In types 1 and 2, there are four columns of pore-pairs; in type 3, the pore-pairs are irregularly triserial; while type 4 resembles type 3, except that near the peristome the pores are central.

(b) the shape of the region changes. In types 1 and 2, the hypophyllodes are slightly expanded and become narrow at the peristome. In type 3, the ambulacra narrow steadily from the ambitus to the peristome. In type 4, the ambulacra are similar, but expand at the peristomial margin.

Since such a morphogenetic sequence may throw light on the problem of determining the age of the fossil form A. aff. recens, it is unnecessary to offer excuses for recording further information concerning the European species on which the sequence is in part founded. The specimens of A. recens in the British Museum (Natural History), mentioned above, and examples of Trematopygus faringdonensis both in the same Museum, and in the Sedgwick Museum, Cambridge, differ in some respects from the figures given by Hawkins. In the specimens of T. faringdonensis which I have examined, the pore-pairs in each hypophyllode^† are arranged in four columns (text-figure 13) as in types 1 and 2. In Apatopygus (type 4), there is an increase in the number of plates near the peristome which bear central pores; this is probably connected with the more marked invagination of the peristome in Apatopygus, as compared with that in Trematopygus. The width of the paired ambulacra is less in the region of these plates than in the region bearing four columns of pores. Above these plates, the pores are arranged as in type 2 in ambulacra 1 and 5 but all trace of phyllodal characters is lost in ambulacrum 3. The evidence of these specimens thus strengthens the hypothesis of a morphogentic sequence.

One of the changes seen in this sequence occurs in the extra-petaloid parts of the ambulacra; the pore-pairs of types 1–3 (Lower Oolite to Cretaceous) are replaced by single pores in type 4 (Recent). Lambert has described a species of the related genus