ferred Under the new Rules, however, an original spelling is valid when it reveals no inadvertent error (not including an error of transliterating into the Latin alphabet). The generic name Enicocephalus and the family-group names Enicocephalidae and Enicocephalinae must therefore be retained.

There is an apparently growing tendency to drop the term “nymph” in favour of the more general term “larva”; this is related to the disfavour into which has fallen Berlese's theory of the origin of insect larvae. However, the word “nymph” still has descriptive value as pertaining to a particular kind of less specialized larva, and since it is still widely employed in systematic and biological work on the Heteroptera, its use has been retained in this series.

To date, only three species of Enicocephalidae, belonging to three genera, have been recorded from New Zealand. The new forms described in this paper now bring the total to eight species and five genera. The composition of the New Zealand fauna is most interesting in several respects. Three of the genera and four of the species belong to the small subfamily Aenictopechinae, a primitive group with a southerly distribution. Outside New Zealand only two species are known certainly to belong to this subfamily, Gamostolus subantarcticus Berg (Patagonia) and Aenictopechys necopinatus Breddin (Java and Sumatra); Cocles contemplalor Bergioth (Madagascar) has been included, but its position is still uncertain (p. 395). Thus about half of the known genera and over half the known species of this archaic group are found in New Zealand, though it seems probable that more representatives remain to be discovered from the less intensively worked areas of the Southern Hemisphere.

Two of the four New Zealand species of Enicocephalinae belong to the monogeneric tribe Phthirocorini, a peculiar and in some respects probably primitive group otherwise known only by one species from the Crozet Is. The other two species belong to the widespread genus Systelloderes Blanchard.

Four-fifths of the genera and three-quarters of the species of New Zealand Enicocephalidae thus belong to two relatively primitive groups of southern affinities. By contrast, less than one-third of the world genera and less than one-eighth of the world species belong to these groups. The predominant and widespread tribe Enicocephalini, which is probably the latest evolved, is not represented in New Zealand at all.

Another interesting feature of the New Zealand fauna, paralleled in other groups of animals, is the unusually high proportion of flightless, apterous or micropterous species. Four of the eight species are of this form, while three of the five genera, so far as known, include only flightless species.

There is a high degree of endemicity. All eight species so far as known are endemic to the Maorian subregion (Phthirocoris magnus occurs in the Auckland Islands as well as in the South Island of New Zealand proper). Three of the five genera are endemic, and four of the eight species belong to endemic genera.

The Enicocephalidae are related to the Reduviidae and have usually been placed in the superfamily Reduvioidea. They comprise, however, a very distinct family which some recent workers have raised to the status of a separate superfamily, the Enicocephaloidea (e g, Poisson, in Grassé, 1951). The present tendency is to regard them as having originated from a very early offshoot of the Heteroptera. The Enicocephalidae differ from all other Heteroptera in the following combination of characters: terrestrial bugs with the body not excessively flattened; predacious, with the rostrum held well away from head at base; antennae and rostrum 4-segmented; head usually divided by a transverse constriction into two lobes (divided in all New Zealand species), the postocular lobe often subglobular and bearing the ocelli near its anterior margin (in the adult, ocelli absent only in Phthirocoris and Nymphocoris, both occurring in New Zealand); pronotum widening towards the base, usually with two transverse constrictions, the anterior one sometimes obscure, the posterior one sometimes obscure or absent (absent or obsolete in the New Zealand genera Nymphocoris

and Aenictocoris); prosternum without a median stridulatory groove; front wings when fully developed entirely membranous (corium lacking) and with the venation extending uninterrupted throughout whole length, in Phthirocoris, Aenictocoris and Nymphocoris wings absent or vestigial and veinless or with few veins; front legs raptorial, more or less incrassate; apex of front tibiae usually dilated, armed with spines, tubercles or plate-like processes and with a pecten or row of close spines along inner (anterior) margin; front tarsi 1- or 2-segmented; middle and hind tarsi 2-segmented, the basal segment very short; tarsi with one or two claws. The Enicocephalids also differ from other Gymnocerata in lacking the paired scent-gland openings of the thoracic venter; the adults instead retain from the nymphal instars the single median opening on the fourth abdominal tergum. Some species show marked sexual dimorphism (in New Zealand, the species of Systelloderes). The family is more fully redescribed and its systematics dealt with by Jeannel (1942) and Usinger (1945).

The nymphs are similar in general structure to the adults, but lack ocelli and fully developed wings, while the eyes are smaller, with fewer ommatidia, and the prothorax, especially the posterior lobe, is reduced and often differently formed. Nymphs are thus readily distinguished from adults of alate species, but are not so obviously different from neotenic, apterous or micropterous adults like those of Phthirocoris, Aenictocoris and Nymphocoris. Such neotenic adults, however, have fully developed terminalia or genitalia (the pygophor and associated structures fully formed in the male; the apical plates of the abdomen sclerctised in the female) and have two-segmented middle and hind tarsi, whereas in nymphs these are always of one segment only.

In all stages the first abdominal segment is reduced, the first sternum small and divided and usually not visible in intact specimens, and the first tergum is without connexiva and separated only by a groove from the second tergum, which is not overlapped by it as are the other terga by their predecessors. The first tergum is thus relatively inconspicuous, and indeed may be covered by the wings in the last instar nymphs of alate species; the second tergum is often referred to in the literature as “the first visible tergite”. However, in all species considered here, the reduced first tergum is present at all stages and clearly visible in the earlier instar nymphs and also in adults when the wings, if present, are spread Where there is any doubt, the segments may be orientated from the fourth tergum, with its median orifice in both nymphs and adult, or by reference to the genital segments (in the female the last complete sternum is the eighth, the last complete tergum the ninth; in the male the tergum and sternum of the ninth segment are fused to form a capsule-like or ring-like pygophor (pygidium)). In the descriptions the abdominal terga and sterna are numbered by reference to their appropriate segments, as indicated above.

Since only preserved material has been available for study, the nymphal instars were assigned to their respective species by a serial comparison with the adult forms, starting with the last instar and working down to progressively earlier stages. As indicated in the descriptions, certain of the distinctive specific characters of the adults are displayed also in the nymphs. Comparison of locality with the known adult distribution was an additional guide in the initial sorting. Instars of any one species are usually readily distinguishable by a combination of size differences (as shown in the usual way, utilising “Dyar's Law”, by measurements of hard parts, such as head and pronotum) and differences in structure and proportions.

Total length is exclusive of the rostrum (which may be extended at varying angles) and thus of the labrum (an elongate triangular sclerite curving above the short first rostral segment); for apterous and brachypterous adults and for nvmphs it is measured to the apex of the abdomen, but in macropterous adults to the apex of the closed fore wings, since in the female the abdomen is very variable in size and