and bluntly rounded; posterior angles rounded, not or scarcely projecting beyond bases of hemelytra, less angled and less prominent than in nasalis, and width between them less in proportion both to median length and to anterior width (between outer bases of anterior processes) (post. W.: L. = 2.1; post. W.: ant. W. = 2.1–2.2); disc between posterior angles rather shallowly convex, main anterior part of disc not forming a plane at a marked angle to posterior part, and whole pronotum thus flatter and more gradually and evenly rounded than in nasalis and less prominently declivous in front. Punctation of ventral surface of metathorax deep but rather fine; adjacent punctures rarely touching, mostly with at least one puncture width between them. Scutellum and corium more finely punctate than in nasalis. Membrane of hemelytra strongly declivous, only very shortly surpassing apex of abdomen Legs proportionately rather shorter than in nasalis (posterior tibia 0.62–0.65 times as long as corium in hudsoni, 0.70–0.76 times in nasalis). Abdomen very finely punctate beneath.

Types: In G. V. Hudson Collection, Dominion Museum, Wellington Holotype ♀ (10u) and 1 paratype ♀ (10v) from Arthur's Pass, Southern Alps, S. Island, 20.1.1940; 1 paratype ♀ (10w) from near Hermitage, Mount Cook, S. Alps, S. Island, 28.1.1945.

Close to Cermatulus nasali nasalis (Westwood), but readily distinguished at first sight by its much more smoothly rounded appearance, both in outline and in surface view, and its smaller size, and, in detail, by the proportionately shorter antennae, by the flatter, less declivous, and more finely punctate pronotum, with the posterior angles more rounded and less projecting and the width between them proportionately less, and by the fine, well-separated punctures of the undersurface of the metathorax.

The chief characters distinguishing the third New Zealand subspecies, turbotti, from nasalis and hudsoni, are given in the key. Through the courtesy of Mr. T. G. Campbell, of the Division of Entomology, C.S.I. R. O., Canberra, the author has been able to examine a specimen of another, as yet unnamed, subspecies from Mt. Rufus, Tasmania It is thought advisable to add here a preliminary list of common features by which all three subspecies already named differ from this alpine Tasmanian form: postero-lateral angles of pronotum more or less rounded, never produced into acute spines; without a distinct transverse ridge just behind calli: spout of metathoacic scent-gland orifice, including posterior edge, thick and much raised above general surface; the granular, impunctate evaporating area in front of spout deeply and sharply transversely grooved, and the evaporating area behind spout large and conspicuous; posterior part of the two ventro-lateral plates formed by the 8th abdominal tergum of the ♀ not strongly deflexed.

Discussion: There is no obviously clear-cut spacial discontinuity between the ranges of nasalis and hudsoni; isolation would seem rather to have arisen through adaptation of the latter to a distinct, subalpine set of habitat factors, topographically determined. (The case of hudsoni thus presents some interesting differences from that of the insular subspecies turbotti (Woodward, 1950, pp. 29–30), in which there are no such obvious differences in habitat from that of nasalis, to which the insect might become adapted, and in which, on the other hand, there is clear-cut geographical isolation. In the evolution of both hudsoni and turbotti, relative smallness of population may have to be taken into account,

but would seem to be of greater importance in the latter, where other factors are not so apparent.)

With a physiological and bionomic adaptation of this kind, it might be expected, though it is not yet known, that there is an intermediate zone in which both subspecies might survive, although presumably not so successfully as in the habitats to which they are respectively more perfectly adapted. The fact that what must be the comparatively small population of hudsoni has maintained its integrity as a structurally distinct form in its present known range, seems to indicate that differentiation has proceeded so far that interbreeding with the much larger adjacent population of nasalis either no longer occurs to any great extent or, if it still does, the hybrids are either sterile or are selectively climinated (possibly involving semigeographic speciation of Mayr). A knowledge of the distributional range of the two subspecies in alpine and subalpine regions, and of their bionomics, physiology, and internal anatomy, is highly desirable, and would no doubt throw light on the mechanisms of speciation involved. In cases like this, while the structural differences may be apparently “neutral” or non-adaptive and perhaps to be accounted for by “genetic drift” in a relatively small population, the possibility must be considered of their being genetically correlated with the adaptive physiological differences that must occur. On the other hand, this is not to exclude the possibility of “drift” and adaptation being involved together, since the isolation of a population by means of physiological and bionomic adaptation might lead subsequently to “drift” in structural characters.

The writer originally described turbotti as a species, but has since had the opportunity of examining a larger series of Cermatulus, including Australian material, and, with a wider experience of the degree and kind of variation involved, now considers that this and the other forms which can be separated from typical nasalis are best regarded as subspecies. So far as is shown by the material yet to hand, these forms are morphologically quite distinct, each being characterized by a group of constantly correlated structural features and readily identifiable from single individuals. The differences between them are, in fact, of the kind which have often been used to separate closely related species of insects, and no doubt some workers will prefer to regard these forms as such. In the absence of any universally applicable and universally acceptable definition of a species and subspecies—which is perhaps not possible—no absolute values can be attached to these terms, and the subspecies or incipient species of one person will be the species of another Both forms are well-established stages in the process of speciation which cannot be sharply delimited.

It seems to the writer desirable to make appropriate use of these two available taxonomic categories as an aid in expressing probable relationships and degrees of divergence, in a similar way to which genera and subgenera can be used at a higher level. In the present instance the most natural grouping appears to be to place the four known nasalis-like forms, which are obviously all closely related and at an early stage of divergence, as subspecies. To refer them to separate species would be to obscure, on the one hand, this relationship and, on the other, the much greater differentiation between the Australo-New Zealand group and pulcher, the New Guinea species. The former interpretation would seem also to accord better with the present-day dynamic trend in systematics, and with the concept of the polytypic species and the geographic race or subspecies, as elaborated by Mayr and others.