Go to National Library of New Zealand Te Puna Mātauranga o Aotearoa
Volume 81, 1953
This text is also available in PDF
(3 MB) Opens in new window
– 221 –

The Fresh-water Mollusca of New Zealand
Part 1—The Genus Hyridella

[Read before Biology Section, Wellington Branch, July 9, 1952; received by Editor, July 11, 1952.]

Abstract

The described forms of Hyridella in New Zealand are considered from the viewpoint of population systematics, many of them being demonstrably ecotypes rather than subspecies. Examination of photographs of the types of Unio aucklandica Gray and Unio Menziesi Gray has assisted in fixing the identity of these two forms morphologically and geographically. Unio lutulentus Gould is shown to be a synonym of Unio aucklandica Gray. Two species, each with two geographic subspecies, are recognised. Analysis is made of the variation in shell proportions of 49 populations of fresh-water mussels.

Introduction

The systematics of the fresh-water molluscs of New Zealand have received scant attention during the last forty years. Suter, in the “Manual of New Zealand Mollusca” (1913) provided the most recent comprehensive treatment. The account given by Suter has been found to be incomplete and difficult to use. Large collections have been amassed in some groups and the systematics require much modification in the light of modern taxonomic theory and practice. The present account deals with the fresh-water mussels. It is hoped to consider other groups in later papers.

The writer wishes to express his thanks to Dr. R. A. Cumber, Mr. A. W. B. Powell, and the late Mr. A. C. O'Connor for the loan or gift of material and for useful discussion. To Dr. W. J. Rees, of the British Museum (Natural History) he is indebted for photographs of the types of Unio aucklandicus Gray and Unio Menziesi Gray, and to the British Museum (Natural History) for permission to publish them. The other photographs used were produced by Mr. C. Hale, of the Dominion Museum.

The systematics of Hyridella in New Zealand have occasioned a good deal of difficulty. New Zealand fresh-water organisms seem particularly plastic and variable. Specific or subspecific limits are difficult to define, and systematic workers encounter many difficulties. Such groups as the Molluscan genera “Isidora”, Lymnaea, and Potamopyrgus and the fish genus Galaxias are notorious in this respect. Similar fresh-water groups overseas have also proved “difficult,” but the situation appears to be accentuated in New Zealand. This state of affairs may, perhaps, be due to fluctuating drainage patterns in the Pleistocene and to the relative youth of present fresh-water systems.

A similar situation pertains in a number of marine and terrestrial groups— e.g., the marine genera Buccinulum, Verconella (Mollusca), and such terrestrial genera as Paryphanta, Placostylus, Phrixgnathus, and Liarea (Mollusca); Melanchra (Lepidoptera) and the plant genus Hebe.

When a large collection of Hyridella from a wide range of localities is examined the range of variation is such that it might well be concluded that only

– 222 –

one very variable species was concerned. When populations are studied rather than single shells some of the problems disappear. Among some populations there is as much variation as there is between two populations widely separated in space. This means that variation and shell morphology is not a simple response to environmental conditions, otherwise all specimens would be similarly affected. Ultimately, nomenclature is a method of expressing relationships, and unless systematic treatment does demonstrate relationships the taxonomist must be prepared to defend the artificiality of his system. By studying populations geographically and ecologically it is hoped that a nomenclature that does express relationships may be evolved.

As an aid to other workers on the group and to assist in evaluating the present conclusions, figures are here presented, as far as possible, of type material of the named forms.

Hyridella Swainson, 1840. Treat. Malac., pp. 285, 380.

Type (monotypy): Unio australis Lamarck.

Synonym: Hyridunio Iredale, 1934. Austral. Zool. 8: 68. Type (o.d.) Hyridella australis drapeta Iredale.

Iredale (1934, p. 67) has shown that there is great difficulty in determining the shell to which Unio australis Lamarck should apply. He has, therefore, introduced a new genus, Hyridunio, to cover the shells previously assigned to Hyridella. He then, however, treats australis as a species of Hyridunio, fixes a type locality as coastal rivers of New South Wales, near Sydney, and makes a sketchily diagnosed subspecies of australis the type of Hyridunio. This means that Hyridella disappears from Australasian faunal lists, although the type species, australis, is included among the species of a new genus. Such an action, although possibly useful in establishing a nomenclature, is not in accord with modern systematic practice and establishes a rather dangerous precedent. The Official Record of the International Commission on Zoological Nomenclature at its session held in Paris in July, 1948, as published in Bull. Zool. Nomenc., vol. 4, pp. 158-159, 1950, recommends that “an author who publishes a name for a genus is, in the absence of evidence to the contrary, to be assumed to have identified correctly the nominal species referred by him to the genus so named.” Where reasonable doubt occurs the case may be submitted to the International Commission. In this particular case Iredale's action in fixing a type locality for australis removes any ground for misapprehension. Hyridella may therefore be retained for those Australasian fresh-water mussels which appear to be congeneric with australis. The generic characters of Hyridunio Iredale as detailed (Iredale, 1934, p. 68) may therefore be accepted as a generic diagnosis of Hyridella. It is possible that the New Zealand fresh-water mussels are not congeneric with australis. The beak characters of australis are not well known. Iredale states they are “eroded, but apparently wrinkled, ridges being decipherable upon the eroded surface.” Cotton and Godfrey (1938, p. 88) state “juvenile with weak umbonal wrinkles which are discontinued at an early stage in the developement and later eroded leaving the adult smooth,” but include Hyridella in family Mutelidae, one of the family characters being “having a nepionic stage represented by a Lasidium, resulting in unsculptured umbos for the adult shell” Until the juvenile shell of Hyridella australis Lamarck is adequately described and figured the generic placing of the New Zealand species will be in doubt.

– 223 –

In all the juvenile New Zealand shells that the writer has seen, the beaks have the same general sculpture of radiating, nodulose ridges. The material examined represents the forms lutulentus, menziesi, aucklandicus, depauperatus, hochstetteri, rugulosus, lucasi, websteri, and zelebori.

When a large number of New Zealand Hyridella comprising adequate population series from a wide ecological and geographical range is examined critically, certain conclusions become apparent:—

1.

A few fairly well marked forms exist.

2.

Long series of intermediate forms link up these “species,” intermediate forms being more numerous than the “stable” ones.

3.

Extreme variation is often exhibited within a single population. There is more variation in respect to hinge characters, for example, in some small populations than in geographical series.

4.

Similar variants occur in widely separated localities.

There are strong reasons for advocating the use of one specific name to cover all the New Zealand fresh-water mussels. Suter (1913, p. 941) in advocating the retention of aucklandicus Gray as a subspecies of menziesi stated, “I think it is more in the interest of science to separate a number of more or less distinct forms which are produced by differences in their environment. Too much lumping does not tend to advance scientific knowledge.” Suter's treatment of the systematics of Hyridella has stood for forty years, but no local conchologist would claim that Suter's treatment of this group had advanced knowledge. Few, if any, would claim that it was now possible to identify the forms accepted by Suter. Modern work on subspeciation has made it unnecessary to give a name to infra-specific forms “to stimulate further research,” although this is still the stock excuse of the “species-monger.” Investigation of infra-specific forms has continued more vigorously, in fact, when no complicated systematic hierarchy exists. Such a “retrograde” step as lumping all the local forms under one specific name would at least emphasize the extreme plasticity of the group in New Zealand.

There are, however, a few forms which may be defined geographically as well as morphologically. In the belief that these are subspecies in the making, names have been retained for them. There are still many gaps in our knowledge of distribution and variation in this field. The fauna of the lakes of the South Island has not been adequately investigated, nor has the area north of Whangaroa. No adequate series collected from various stations of one drainage system exists. Nor is there any adequate bathymetric data from a series of representative lakes. Until such data have been accumulated the systematic position of the fresh-water mussels cannot be adequately decided.

Hyridella aucklandica (Gray).

Unio aucklandicus Gray, 1843. In Dieffenbach, Travels in New Zealand, ii, p. 257.

Unio lutulentus Gould, 1850. Proc. Boston Soc. Nat. Hist., vol. 3, p. 295.

Diplodon (Hyridella) Menziesi Websteri Simpson; Suter, 1913. Man. N.Z. Moll., p. 943 (in part not of Simpson).

Examination of photographs of the type of Gray's aucklandica kindly supplied by Dr. W. J. Rees shows that Suter misapplied this name. The photographs here reproduced by courtesy of the British Museum (Natural History) are of an elongated shell with subparallel dorsal and ventral surfaces. The height/length ratio is 15. The only populations of New Zealand fresh-water mussels which the writer has seen that are at all comparable with these ratios and the photographs

– 224 –

are a series found in areas around and north of Auckland and which the writer had provisionally classed as a subspecies of websteri Simpson (see later). The type locality of aucklandica is “Bay of Islands” which is in keeping with the known distribution of the above-mentioned form. No specimens of this form have been seen with the sinuous ventral and rounded posterior margins of the type but the type is evidently a much eroded shell and the sinuous ventral margin is a common response by mussels living in stony streams. There seems little doubt that the name aucklandica should be applied to these shells and that websteri should be considered as an allopatric subspecies. * The name aucklandica has priority so must become the nominate form. Suter reduced websteri to the status of a subspecies of menziesi but forms of menziesi and aucklandica aucklandica and aucklandica websteri are sympatric throughout the known range of the latter two forms. Hyridella aucklandica has a rather restricted range with a number of sporadic peripheral occurrences. One or other of the subspecies occurs in association with a form of menziesi at Lake Kaitoke, Wanganui, and at Kaikohe. In other cases full collection data is not available though Suter remarks of websteri that this species “is always found together with aucklandicus” (meaning in this case a form of menziesi). Suter's records of websteri, as examination of his collection shows, refer to both the typical form with lachrymose nodulous sculpture and the smooth form (here recognised as true aucklandica).”

Examination of a cotype of Gould's Unio lutulentus, in the Suter collection (here figured, Pl. 19, Fig. 7) proves that it is indistinguishable from forms of aucklandica aucklandica (Gray). Gould cited the habitat as:—“Inhabits New Zealand, common. Drayton.” Study of the official Narrative of the United States Exploring Expedition reveals that Drayton was an artist attached to the “U.S.S. Vincennes” and that his opportunities for collecting in the New Zealand area were confined to the general area of the Bay of Islands. It would therefore appear that not only is Unio lutulentus Gould identical with Unio aucklandicus Gray morphologically but that the type localities of both species is essentially the same. Suter, in his collection identified very few shells as lutulentus and none of these agree with the cotype of this species. The shells called “lutulentus” by Suter are in fact the low form of menziesi found in streams, and will be discussed later. The cotype of lutulentus does agree quite well with the shell in the Suter Collection from Pukekohe classified by him as websteri, and here recognised as Hyridella aucklandica aucklandica Gray).

Hyridella aucklandica aucklandica Gray. Pl. 17, figs. 2, 5, 6, 8; Pl. 19, figs. 2,3,4,8.

Shell elongate with subparallel dorsal and ventral margins. Similar to the better characterized websteri but with the nodulose sculpture lacking or but poorly developed, and with the concentric sculpture of impressed growth marks accentuated. There is also a size difference, aucklandica aucklandica growing to a consistently larger size than aucklandica websteri. The hinge development varies a good deal. The laterals are fairly consistent, comprising a long lamellar lateral in the right valve and two fine curved laterals in the left valve. In the right

[Footnote] * The terms allopatric and sympatic are used in this paper as defined by Mayr (1942, pp. 148-9). “Two forms or species are sympatric, if they occur together, that is, it their areas of distribution overlap or coincide. Two forms (or species) are allopatric, if they do not occur together, that is, if they exclude each other geographically.

– 225 –

valve there are two pseudocardinals, the anterior small, thin and straight, the posterior being stronger, triangular, with a corrugated edge. In the left valve there is usually a long, oblique pseudocardinal, with at times the trace of a small posterior accessory pseudocardinal. In some cases the original pseudocardinal is divided, apparently because of pressure by the main pseudocardinal in the right valve during development. This can give the appearance of two well developed pseudocardinals in the left valve. In extreme cases the teeth are reduced to formless bosses with corrugated and roughened articulatory surfaces in each valve. At times the degeneration of structure can simulate a hinge of a very different pattern. This is particularly noticeable in the series from Kaeo.

Localities. Bay of Islands (type); Pukekohe (Suter Collection); Creek at Henderson, Auckland (Suter Collection); Creek near Avondale, Auckland (Auckland Museum); Creek near Glen Eden, Auckland P. Parr, 1925, together with menziesi (Auckland Museum); Kaikohe, Northland (R. A. Cumber and R. L. Oliver, —/2/40); Kaeo, Whangaroa (Mrs. I. Worthy).

Hyridella aucklandica websteri (Simpson, 1902). Pl. 17, figs. 1, 3; Pl. 19, fig. 1.

Diplodon websteri Simpson, 1902 Nautilus, 16, p. 30.

Diplodon (Hyridella) Menziesi websteri Simpson; Suter, 1913. Man. N.Z. Moll., p. 943.

The distinguishing features are the subparallel dorsal and ventral margins, and the strongly lachrymose nodulation. Concentric impressed growth marks are well developed, there is a well marked sinuosity in the ventral margin submedianly and often a shallow depression running from the beaks obliquely backward to the ventral margin. No well preserved beaks have been seen but there appear to be at least three radiating nodulose ridges across the juvenile shell. The hinge development varies a good deal as in the nominate subspecies.

Localities. Waiuku (type); “Auckland” (Dom. Mus.); Hamilton, Waikato (Dom. Mus. ex Bollons Coll.); railway cutting, 10 miles from Wanganui, coll. Kirk, 1875 (O'Connor Coll.), one subfossil shell with patches of epidermis still adhering; Lake Kaitoke, Wanganui (A. C. O'Connor); Maori midden, Kerekere Rd., between Paiaka and Foxton, one complete shell, R. A. Cumber—/7/51, and one fragment A. C. O'Connor, 22/8/51. The localities “near Pukekohe and Henderson” given by Suter refer to aucklandica aucklandica.

There are several reasons for considering aucklandica an older form than menziesi. The latter is a widespread form throughout New Zealand, and is presumably an energetic coloniser of fresh habitats. On the other hand, aucklandica has a relatively restricted range, and within this range two well characterized geographic subspecies have arisen. Its distribution pattern is scattered, and it often occurs in association with forms of menziesi, though in most cases in much smaller numbers than the latter. These are all strong arguments for considering aucklandica an older form than menziesi, and that it is gradually being replaced by the latter. The reverse argument that aucklandica is a newly evolved form, may follow from some of the above evidence, but the fact that two subspecies have evolved in restricted areas, the scattered distribution and the occurrence of subfossil forms would seem to disprove such a hypothesis. There does not seem much chance that aucklandica is an ecological form since two distinct species occur in the same area, since aucklandica is found in a variety of habitats (drains, streams, and lakes) and since its distribution is relatively restricted. It is therefore presumed that aucklandica in the past had a somewhat wider distribution

– 226 –

than at present, but was restricted to the general area of the North Auckland peninsula south of Whangaroa, and to an area south of Auckland bounded on the east by the old course of the Waikato to the Firth of Thames, spreading south to the vicinity of Foxton. Some barrier, probably sea, divided the population into two isolated portions. Such a barrier might well have been in the vicinity of the Auckland isthmus. In comparative isolation two subspecific forms have developed, aucklandica websteri in the South, auckclandica aucklancdica in the North. Re-invasion of the range of aucklandica websteri by aucklandica aucklandica into the South Auckland area as far as Pukekohe with the extinction of the species in numerous intermediate localities due to competition with the more adaptable menziesi stock would give the present distribution pattern.

Hyridella aucklandica aucklandica (Gray).
Measurements and indices. (All specimens in Dominion Museum.)*
Locality A B C D E F G
Kaeo 94 44 47 20 21 19 20
87 41 51 24 28 17 19.5
88 43 48 20.5 23 19 21
94 44 47 21.5 23 16 17
90 43 48 20.5 23 16 18
89 43 48 20 22 16 18
80 37.5 47 19 24 14 18
75 37 49 15 20 16 21
Pukekohe 56 28 50 12.5 22 11 20
Henderson 78 38 49 17.5 22 14 18
72 35.5 49 15 21 15 21
Kaikohe 70 33 47 16 23 13 19
75 30 40 16 21 10 13
68 32 47 14.5 21 12 18
67 32 48 15 22 11 16
68 32 47 14.5 21 12 18
67 31 46 16 24 10 15
67 31 46 14 21 12 18
Bay of Islands (type) 45 18
Ranges and Means. (Means in parentheses.)
Length 56-94(77) mm.
Height Index 46-50(47)
Width Index 20-24(22)
Beak Index 13-21(18)
Hyridella aucklandica websteri Simpson.
Measurements and indices. (All specimens in Dominion Museum.)
Locality A B C D E F G
Waiuku (co-type) 60 32 53 16 27 14 23
Waiuku (paratype) 69 33 48 17.5 25 18 26
Hamilton, Waikato 51 26 51 12 24 10 19.5
Auckland 54 28 52 14 26 12 22
Kaitoke Lake, Wanganui 58 28 48 12 21 11 19
Railway Cutting, Wanganui 77 38 49 16 21 15 19.5
Foxton 62 30 49 12 19 13 21

[Footnote] * The measurements and indices used in this paper are modifications of those used by Suter (1905, p. 241). A is the length, B the height, C the Height Index—i.e., the height expressed as a percentage of the length, D is the width, E the Width Index—i.e., the width expressed as a percentage of the length, F is the distance from the beak to the anterior end of the shell and G is the Beak Index—i.e, the distance from the beak to the anterior end expressed as a percentage of the total length. The measurements were taken as shown in Text-Fig. 1.

– 227 –
Ranges and Means. (Means in parentheses.)
Length 51-77(62) mm.
Height Index 48-53(50)
Width Index 21-27(23)
Beak Index 19-26(21)
  • “Hyridella lessoni” Suter 1905 (non Kuster, 1856). Pl. 17, fig. 4.

  • Diplodon lessoni(Kuster 1856); Suter, 1905. Trans. N.Z. Inst., vol. 37, p. 240. (non Kuster).

  • Diplodon (Hyridella) Lessoni Kuster; Suter, 1913. Man. N.Z. Moll., p. 938 (non Kuster).

Suter sent specimens of Hyridella from 10 to 30 feet in Lake Wakatipu to the U.S. National Museum where Dall identified them as Diplodon lessoni Kuster, a species originally described from New South Wales. Suter (1905, p. 240) expressed some doubt but accepted the identification. Iredale (1934, p. 73) has shown that Kuster's lessoni was a substitute name for depressa Lesson, 1831 (non Lamarck, 1819) and that the correct name for the New South Wales shell is Propehyridella nepeanensis (Conrad). The New Zealand shell is totally unlike nepeanensis, judging by the figures given of that species by Iredale (1934) and Cotton and Gabriel (1932). The strong radials which are a feature of nepeanensis are totally lacking in the New Zealand “lessoni” while shell outlines are quite different. The name lessoni must therefore be removed from the New Zealand list. Four shells, including the specimen figured by Suter, remain in the Suter Collection. In outline and general characters they appear so close to undoubted forms of menziesi that there is no need to rename the Wakatipu shells (Text figure 1).

Picture icon

Text-Fig. 1.—Outline drawings of: A, Hyridella menziesi menzies Gray, Henderson; B, C, Hyridella” lessoni” Kuster, Lake Wakatipu. (C, is the specimen figured by Suter.) D, Method of measuring Hyridella in this work. O-Q, length; O-P, distance from beak to anterior margin; O-R, height; S-T, diameter.

– 228 –

Hyridella menziesi Gray.

This wide ranging, variable species has been divided into a number of so-called “subspecies,” the most recent grouping being that of Suter (1913). Subspecies as used by Suter in this and in most other groups are varieties rather than geographic, ecological or physiological subspecies. The results of the study of speciation had not influenced systematic work in Suter's time and the reality of subspecific forms with restricted geographic ranges was not realised. Though Suter's subspecies are really varieties, the systematic names used were all validly proposed, and must be considered. The varietal rather than subspecific standing of Suter's forms of menziesi is well shown by the discontinuous and often overlapping ranges given for such groups as menziesi menziesi, “aucklandicus,” hochstetteri and rugatus (= rugulosum). What is of course apparent, is that these varietal forms crop up throughout New Zealand wherever similar limnological conditions occur. Similar situations have been recorded for European and North American mussels, by Ortmann (1920), Bull (1922), Agnell (1949) etc., and a summary of such work has been given by Eagar (1949). Such a nomenclature does not illustrate the natural relationships of the named forms. Subspecies A (from the River Avon) is related to subspecies B which occurs elsewhere in the same river system, rather than to subspecies A from Lake Taupo. Where a varietal form has become stabilised in one general area so that a comparatively restricted area comprises its geographical range, the use of such varietal names in a subspecific sense may be warranted. Such a form as depauperatus, confined to lakes in the vicinity of Auckland and Northland, should probably be treated as subspecific. Such forms as crop up in widely separated areas under similar conditions do not deserve taxonomic recognition. In the past such forms have been named, either as part of a species-mongering campaign or “to direct interest to the problem.” Such practices have only served to bring the study of systematics into disrepute and to distract the worker on distribution. Study of such forms outside the systematic structure may, however, be of great value as has been shown overseas. They can be indicated by a non-systematic term and generalisations derived from such study may be of great palaeontological value (Eagar, 1948).

The treatment of the subspecies of Hyridella menziesi will follow these lines.

Hyridella menziesi menziesi (Gray, 1843). Pl. 18, figs. 2, 3, 4, 5, 7; Pl. 19, figs. 5, 6.

  • Diplodon Menziesi Gray, 1843. in Dieffenbach, Travels in N. Z. p. 257.

  • Unio waikarense Colenso, 1845. Tasm. Journ. Nat. Sci., vol. 2, p. 250.

  • Unio Hochstetteri Dunker, 1862. Malak. Blatt., vol. 8, p. 153.

  • Unio Zelebori Dunker, 1866. Verhandl. Zool. -bot. Gesellsch. Wien. vol. 16, p. 915.

  • Unis (sic) rugatus Hutton, 1883. N. Z. Journ. Sci., vol. 1, p. 478.

  • Diplodon menziesi lucasi Suter, 1905. Trans. N.Z. Inst., vol. 37, p. 239.

  • Diplodon Menziesi Gray; Suter, 1913. Man. N.Z. Moll., p. 940.

  • Hyridella menziesi aucklandica Suter, 1913. Man. N.Z. Moll., p. 941 (not of Gray).

  • Diplodon menziesi rugulosum Simpson, 1914. Cat. Naiades, p. 1291 n.nom. for rugatus Hutton (non. Menke, 1828; non. Rossmassler, 1837).

The photograph of the type kindly provided by Dr. W. J. Rees and reproduced here (Plate 18, figs. 2, 7) shows the characters of the nominate form of menziesi. It is quite obviously the high winged shell usually found in lakes. In Gray's original description the locality was given: “Inhab. New Zealand. Rivers in the N. Island and Lake Taupo. Dr. Dieffenbach.” The type of Unio menziesi in the

– 229 –

British Museum apparently bears no locality other than “New Zealand.” The type locality of menziesi has therefore been unknown. Fortunately the type shell is very high relatively, the height/length ratio calculated from the photograph being 68. Such a high ratio occurs in very few populations—e.g., in the Northland Lakes, Lake Rotorua, River Avon, Lake Rotoiti, Lake Letitia, Lake Pearson, Lake Taupo, and in one Wairarapa population. Eliminating the South Island localities which are not indicated by Gray, the central lakes of the North Island would appear to be the most likely provenance of the type of menziesi. In a population from Lake Taupo itself (mentioned by Gray) specimens which match the type of menziesi may be found, and Lake Taupo is therefore here selected as the type locality for Unio menziesi Gray. In dealing with fresh water mussels a type locality is even more necessary than usual, as it is the characters of a population that must be considered rather than these of a single individual.

The characters of a number of populations from Lake Taupo may be summarized here as exemplifying the characters of typical menziesi:—

Motuopa, Lake Taupo (10 specimens); length, 52–77 mm.; height, 39–50 mm.; height/length ratio, 60–77 (mean 69); width, 15–25 mm.; width/length ratio, 25–38 (mean 31).

Lake Taupo, 10-30 feet (Suter Collection, 3 specimens); length, 48-63; height, 32–41 mm.; height/length ratio, 65–69 (mean 67); width, 16–22 mm.; width/length ratio, 32–35 (mean 33).

Lake Taupo, up to 100 feet (Suter Collection, 2 specimens); length, 37–43 mm.; height, 24-28 mm.; height/length ratio, 65–67; width, 9.5–12 mm.; width/length ratio, 26–28.

Such high winged, oval shells are usually found in lakes and ponds throughout New Zealand.

  • Hyridella menziesi hochstetteri” (Dunker, 1862). Pl. 18, fig. 3.

  • Unio Hochstetteri Dunker, 1862. Malak, Blatt., vol. 8, p. 153.

  • Diplodon menziesi hochstetteri Dunker; Suter, 1905, Trans. N.Z. Inst., vol. 37, p. 237.

  • Diplodon (Hyridella) Menziesi Hochstetteri Dunker; Suter, 1913. Man N.Z. Moll., p. 942.

Originally described by Dunker from specimens collected in the Waikato River by Hochstetter, this name has been applied by Suter to any form showing posterior truncation. In this sense it is the “pathological subspecies” of Suter (1905, p. 237). Practically any form of menziesi, “lutulentus” or “zelebori” can develop this posterior truncation, which may be due to the physical nature of the substratum, the chemical composition of the water, or to the attacks of some parasite. In Lake Rotorua such a form is the common one. Non-eroded, normally developed shells do occur, however, and a series from Ohinemutu in the O'Connor Collection show the features of the normal shell. These specimens are broadly oval and high in relation to the length. The truncated form from other Lake Rotorua localities would appear to be derived from this type of shell. Dunker's type came from the Waikato River. The writer has examined small populations from Huka Falls northwards, and has found nothing to match forms from Lake Rotorua. The exact application of the name hochstetteri must wait until the type can be examined. However, normal populations from Lake Rotorua are so close to Lake Taupo specimens that the distinction as used by Suter cannot be maintained. In fact, it seems most probable that hochstetteri

– 230 –

is a complete synonym of menzies and cannot even be applied in a subspecific sense. Suter's records of hochstetteri from the Kapuarangi River, River Avon, and Cheviot apply to truncated shells of the local populations.

Hyridella menziesi lucasi” (Suter, 1905). Pl. 17, fig. 7.

  • Diplodon menziesi lucasi Suter, 1905. Trans. N.Z. Inst., vol. 37, p. 239, figs. 2, 3.

  • Diplodon (Hyridella) menziesi Lucasi Suter, 1913. Man. N.Z. Moll., p. 942.

This subspecies was based upon three shells from 60 feet in Lake Manapouri. The largest specimen is only 45 mm. long, the other two being 34mm. and 20 mm. (not 28 mm. as given by Suter, 1905, p. 241) respectively.

The species was differentiated from aucklandica “by its exceptionally compressed form, the thinness of the shell, the strongly marked and close concentric lines, the more tapering posterior margin, and the feebly developed pseudo cardinals.” The two juvenile shells do not agree with the type having a higher posterior dorsal outline, and falling well within the range of variation of menziesi. The compressed, thin shell of the type and the feeble development of the hinge are all compatible with the deep water habitat which leaves the shape and form of the sculpture as the major differences. The two juvenile shells show that the shape is not a constant feature. No additional specimens appear to have been collected from Lake Manapouri since 1902, so that the three original specimens still remain the only available population sample from this area. Until topotypic material is collected it seems best to drop the name lucasi, at least for the present. Three variable non-adult shells do not supply sufficient material for the erection of a subspecies of fresh-water mussels.

Hyridella menziesi rugulosus” (Simpson, 1914).

  • Unis (sic) rugatus Hutton, 1883. N.Z. Journ. Sci., vol. 1, p. 478.

  • Unio rugatus Hutton, 1884. Trans. N.Z. Inst., vol. 16, p. 216.

  • Diplodon menziesi rugata Hutton; Suter, 1905. Trans. N.Z. Inst., vol. 37, p. 238.

  • Diplodon (Hyridella) menziesi rugata Hutton; Suter, 1913. Man. N.Z. Moll., p. 943.

  • Diplodon menziesi rugulosum Simpson, 1914. Cat. Naiades, p. 1291. n. nom. for rugatus Hutton (non Menke, 1828; non Rossmassler, 1837).

  • Hyridella menziesi rugulosus Simpson; Powell, 1946. The Shellfish of New Zealand, p. 60.

Originally diagnosed by Hutton from Lake Pearson, rugatus is almost certainly an ecotype. The main distinguishing features have been taken to be the thin shell and the rough, irregular concentric striations. Suter applied the name to a series of shells from Canterbury lakes, but included specimens from Lake Kanieri; Wairau River; and Awaka Stream, Clutha, in the South Island, and from Lake Waikare and the Kapuaronga River. Such a distribution pattern immediately renders rugatus suspect as a true geographical subspecies. It would appear that the “rugatus” form occurs in certain lakes where mussels develop thin shells with deep, irregular growth corrugations. The name based upon a Lake Pearson population would, however, be available in a true subspecific sense if it is ascertained that the mussels of the lakes of the eastern coast of the South Island form a morphological entity. The writer has examined specimens from small lagoons off the Pelorous River, and from Lakes Pearson, Letitia, Coleridge, Wanaka, Wakatipu, Te Anau and Manapouri. Although there is general conformity to an elongatae, oval shape and other characters with lack of markedly winged forms, the range of variation is too wide to allow the use of rugulosus as a subspecific name for this series. Later more intensive work may show that this course is possible. In the meantime, rugulosus can have no taxonomic application.

– 231 –

Hyridella menziesi depauperata (Hutton, 1883). Pl. 17, figs. 9, 10; Pl. 18, figs. 1, 8.

  • Unis (sic) depauperatus Hutton, 1883. N.Z. Journ. Sci., vol. 1, p. 478.

  • Unio depauperatus Hutton, 1884. Trans. N.Z. Inst., vol. 16, p. 216.

  • Diplodon Menziesi acuta Suter, 1907. Proc. Mal. Soc., vol. 7, p. 239.

  • Diplodon Menziesi acutus Suter, 1913. Man. N.Z. Moll., p. 940.

  • Diplodon (Hyridella) Menziesi depauperatus Hutton; Suter, 1913. Man. N.Z. Moll., p. 941.

This is an undoubted ecological form of menziesi, with a restricted geographical range. The species was described from Lake Takapuna, Auckland. There are two syntypes in the Dominion Museum, the better-preserved shell here being selected as lectotype (Pl.—, Fig. 8). A single type specimen has little applicability in the study of fresh-water mussels. Series from the type locality show that there is fair uniformity in shell outline, shape and texture, although quite considerable variation in the height and development of the posterior keel. There is a fair degree of agreement in lacustrine forms from other localities north of Auckland—e.g., sand-dune lakes at Muriwai; Lake Crosland, near Helensville; Kaipara Lake; and Tangonge Lake, Kaitaia. The extremely thin shell and the shape are characteristic features.

The systematic standing of acutus Suter from Lake Omapere is difficult to decide. The typical form is readily distinguishable from depauperata, though obviously derived from it, but the range of variation present in a topotypic series renders the discrimination of populations of acutus and depauperata difficult. The distinctive shape of “typical” acutus does occur in other thick-shelled populations of menziesi—e.g., Lake Waikare. Suter (1907, p. 239) differentiated acutus as follows:—

“The outline of this subspecies is very nearly that of D. depauperata, Hutton, but the hinge is very different … the hinge (of acutus) not different from the type.” The hinge in topotypic shells of depauperata is very weakly developed, but it differs from that found in menziesi in degree of development. The whole nature of depauperata lies in an attenuation of the limy layer of the shell. Depauperate shells from other localities (e.g., Muriwai lakes) have a strong hinge though the shell is still referable to depauperata.

If taxonomy is to express relationship acutus must rank as a sub-subspecies of depauperata. There is no such category available to systematists, nor does there seem to be any necessity for adopting such a system. Carried to its logical limits, the naming of infra subspecific categories would result in the separation of practically every population of variable groups within general like Hyridella or the land snail genus Paryphanta. In this particular case the variability of the Lake Omapere population renders its recognition rather unnecesary. The Lake Omapere population may be referred to as the “acutus” form of Hyridella menziesi depauperata without encumbering the systematic nomenclature further. Later workers may find it necessary to re-establish acutus though the status of such a form must always remain a matter of opinion. It may well be that acutus is a subspecies in the making.

Impoverishment of limy material has been noted in shells from lacustrine habitats elsewhere. It is not general in New Zealand lake forms of Hyridella, nor is it usually so well marked as in Hyridella menziesi depauperata. The cause may well lie in a lack of available calcium carbonate.

– 232 –

The forms of menziesi already considered have been in the main from lentic habitats. Specimens from running waters have previously been classified under the three names lutulentus Gould, zelebori Dunker, and “aucklandica” Gray. It has already been demonstrated that Suter's use of aucklandica for these forms was incorrect, so this name is no longer available in this sense. The form which Suter (following Reeve) called by this name is the shell from streams and rivers in which the dorsal and ventral margins are subparallel with a slight wing posteriorly. There is little if any geographical correlation between the occurrences of this form and intermediates between it and typical menziesi abound. Suter vacillated between separation and union of the two forms. There seems no need to institute a new name for the form Suter identified as aucklandica.

It has already been demonstrated that Unio lutulentus Gould is an absolute synonym of Unio aucklandicus Gray. Suter must have accepted lutulentus with reserve as he has specimens in his collection from only two localities. In a series from the Otana River, a confluent of the Wairima, Hokianga, the range of percentages of height and length for ten specimens were found to be 50-63%, average 58%. Collingwood populations have a similar range, 53-64%, average 59%. The equivalent figure for the type of lutulentus is 49%, and for “aucklandicus” as recognised by Suter, 53%. In other respects, too, these shells labelled lutulentus by Suter are best considered ecotypes of menziesi.

As will be shown later when means of populations are considered, all such forms appear as units in a series and there are no clear-cut dividing lines by which these low-winged form with sub-parallel margins may be separated off. The “lutulentus” form is but one extreme of this series with typical menziesi as the other.

In a situation of this kind two treatments are possible. Either the two extremes can be labelled as species and the far larger number of intermediates arbitrarily assigned to one or the other form. Or all may be classified as menziesi. The writer advocates the latter course, at least until an intensive study of the forms of fresh water mussels from one river system has shown the part that ecological conditions can play in modifying morphology.

The shell which has been known as zelebori is a fairly constant form over much of its range, and fills a relatively restricted ecological niche. It shows considerable affinity with lotic forms of menziesi, and may well have been derived from this source. The shell is elongate, with subparallel dorsal and ventral outlines, often with a distinct shallow ventral median constriction. The shell layer is thick, surface usually very clean, with finely developed growth marks on the periostracum. Whether this form is actually the same as zelebori of Dunker is difficult to determine without access to the type specimens. As Suter noted, Dunker wrote of the shell as “tenuicula” and the figure given in the Novara report does not show the ventral constriction. Once again no type locality was given. Some of the shells attributed to zelebori do not possess the ventral constriction, and the figure looks that of a solid shell.

The localities from which this form has been recorded comprise river systems on the east coast of the North Island from the Urewera and Napier south to the Wairarapa. In the South Island it is known from Ashburton, while similar shell types are known from North Canterbury and the Waiau River. While some populations are fairly constant, others show variations approaching closely to

– 233 –

menziesi. Since this form appears to be no more than a stage in the lotic to lentic series of menziesi, there does not seem any good reason for recognising it systematically.

Vriation in the New Zealand Forms of Hyridella

In an attempt to express the variation present in the fresh-water mussels, a scheme of measurement modified from that used by Suter (1905, p. 241) was applied to a large series of shells. From these measurements three indices were calculated for each shell. The measurements were taken as shown in Text-Fig. 1.

1.

Total length.

2.

Height.

3.

Diameter (or obesity), the thickness of the paired valves.

4.

Distance of beak from the anterior end. This is the least satisfactory measurement as the beak is usually eroded, and the exact point cannot be determined.

The indices used were:—

  • (a) The height as a percentage of the length. (Height Index.)

  • (b) Diameter as a percentage of the length. (Width Index.)

  • (c) Distance from beaks to anterior end as a percentage of the length. (Beak Index.)

These indices were calculated for a sample (ten specimens where possible) of each population, and the ranges and means for each population then obtained. The results of these measurements are given in tabular form. These means were then plotted in histograms (Text-figs. 2, 4, 5). It is a matter of some considerable theoretical interest that the three curves for the Height Index, Width Index and Beak Index for all available populations of Hyridella make a very close approximation to the normal curve of distribution. Considering the height/length ratio alone, the conclusion that all the New Zealand forms of Hyridella are very closely related appears inescapable. It is of interest that the only form that has been

Picture icon

Text-Fig. 2.—Histogram of means of height/length ratios for 49 populations of Hyridella.

– 234 –
Picture icon

Text Fig. 3.—Analysis of populations making up height/length ratio in Text-Fig. 4.
Text-Fig. 4.—Means of width/length ratios for 49 populations.

accepted earlier in this work as being definable by shell outline and geographical distribution (i.e., the two subspecies of aucklandica) occupies an extreme of the curve. Even if depauperata were clearly defined it would be more apparent in a weight/length ratio. In Text-Fig. 3, the positions on the histogram filled by various recognisable forms or geographical series have been indicated. It would appear from this that there is little possibility of dividing the menziesi complex further with any degree of validity. Nor does there appear to be any correlation with latitude.

– 235 –
Picture icon

Text Fig. 5.—Means of beak indices for 49 populations.

At attempt was made to correlate height/length ratios with ecological stations. Unfortunately, ecological data is all too often lacking in collections, and the number of stations is therefore too small to be anything but indicative of possible trends. Treated on a broad basis, the figures are:—

Streams and rivers (17 stations). range of height/length index 47-66, mean 57. Lagoons, small ponds, dams, etc. (3 stations), range 58-61, mean 59. Lakes (18 stations), range 57-74, mean 63. These figures would indicate an increase in height as compared with length from lotic to lentic habitats. This problem will require much more intensive investigation before generalisations may be formulated.

There is a highly obese form developed at Kaihoka Lakes, West Haven, the mean obesity of which is slightly discontinuous, at least as far as populations sampled in the present work would show. The shells are, however, always considerably eroded, and there seems no reason to suppose that this form is anything but an ecotype.

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Length H/L × 100 ob/L × 100 Bk/L × 100
Kaihoka, West Haven (7) 75-91 53-68 (59) 32-45 (38) 18.5-26 (22)
Lake Wanaka (4) 77-89 57-63 (61) 29-33 (31) 20-27 (24)
Karaka Lake, Kaipara Heads (8) 55-78 65-69 (67) 19-29 (26) 17-27 (23)
Tuakau, Waikato River (10) 55-65 60-66 (63) 25-32 (28) 20-27 (23)
H. aucklandica websteri (6) 51-77 48-53 (50) 21-27 (24) 19-26 (22)
H. aucklandica aucklandica (18) 56-94 46-50 (47) 20-24 (22) 13-21 (18)
Motuopa, Taupo (10) 52-77 60-77 (69) 25-38 (31) 15-25 (20)
Ohinemutu, L. Rotoiua (9) 56-71 64-72 (68) 32-43 (35) 21-34 (30)
Lake Rotorua (distorted) (10) 35-67 67-78 (73) 24-43 (32) 19-34 (27)
Te Hopai, Waiiarapa (8) 51-61 60-68 (65) 27-31 (28) 18-24 (21)
– 236 –

[The section below cannot be correctly rendered as it contains complex formatting. See the image of the page for a more accurate rendering.]

Kaitoke Lake (9) 52-65 61-67 (63) 30-35 (32) 20-23 (21)
Silverstream (8) 42-49 59-67 (64) 24-31 (27) 20-29 (25)
Kaawa Creek (decorticated) (10) 32-51 58-67 (63) 26-51 (32) 20-28 (23)
Muriwai (10) 49-79 58-68 (63) 27-30 (28) 18-24 (20)
Porirua (10) 60-103 50-62 (57) 22-30 (27) 19-25 (22)
Turanganui Lagoon (10) 63-80 57-67 (61) 26-41 (29) 24-36 (26)
Oyster Shell Creek (10) 53-64 50-65 (59) 27-33 (29) 18-30 (23)
R. Avon, Barbadoes Creek (4) 85-91 53-59 (58) 22-37 (30) 18-25 (22)
R. Avon (6) 58-73 61-70 (66) 28-35 (31) 19-25 (22)
Te Anau (5) 48-60 60-63 (62) 27-31 (30) 23-27 (24)
Lake Kanieri (4) 53-64 59-66 (62) 27-33 (29) 20-25 (23)
Greymouth (2) 69-89 53-64 (58) 25-26 (26) 20-22 (21)
Lake Coleridge (5) 61-84 54-62 (58) 23-28 (26) 16-23 (20)
Lake Hatuma, Waipokurau (6) 54-72 57-65 (61) 27-38 (30) 21-24 (22)
Main Stream, Whangarei (10) 48-60 53-62 (57) 23-28 (25) 17-20 (18)
Feilding (6) 68-77 52-54 (53) 28-32 (30) 22-23 (22)
Waikaremoana (9) 54-69 53-60 (57) 26-30 (28) 16-22 (20)
Napier (Colenso) (3) 60-79 51-56 (53) 28-31 (29) 13-19 (16)
Waiau R. mouth (2) 79-88 58 58 (28-29) 20-28 20-22 (21)
Albury Creek (12) 52-85 51-60 (56) 27-33 (29) 12-20 (15)
Te Papuni (5) 80-102 48-57 (52) 27-30 (29) 16-20 (18)
Ashburton R. (3) 50-83 53-62 (57) 24-28 (26) 16-18 (17)
Napier (2) 68-78 52-54 (53) 28-30 (29) 13-18 (15)
Bainham (4) 46-76 50-54 (52) 25-29 (28) 17-24 (21)
Cheviot (2) 52-55 65-67 (66) 17 (17)
Waimangaroa Creek, Westport (1) 70 (56) (26) (19)
Lake Rotoiti (5) 38-49 70-78 (74) 28-40 (31) 19-27 (23)
Lake Omapere (8) 36-80 56-65 (61) 20-27 (24) 15-20 (17)
Lake Takapuna (10) 44-69 55-64 (59) 22-25 (24) 18-25 (19)
Lagoon of Pelorous River (2) 70 57-61 (59) 24-31 (27) 13-21 (17)
Lake Wakatipu (“lessoni”) (2) 40-51 59-62 (60) 27-32 (29) 22 (22)
Lake Letitia (4) 59-79 57-69 (61) 26-33 (29) 16-24 (20)
Lake Taupo, up to 100ft (2) 37-43 65-67 (66) 26-28 (27) 19-21 (20)
Lake Taupo, 10-30ft (3) 48-63 65-69 (67) 32-35 (33) 17-24 (20)
Lake Manapouri (3) 21-45 55-65 (59) 18-22 (30) 17-24 (21)
“lutulentus” (cotype) (1) 61 (51) (22) (20)
Lake Pearson (“rugatus”) (1) 52 (69) (31) (29)
Wahaou, N. Canterbury (4) 67-75 53-57 (55) 28-31 (29) 17-19 (18)
Collingwood (10) 53-84 53-64 (59)
Hokianga (10) 35-60 50-63 (58)

The forms of fresh water mussels recognised in the present work are as follows:—

Hyridella aucklandica aucklandica (Gray, 1843). Scattered occurrences north of Auckland.

Hyridella aucklandica websteri (Simpson, 1902) Scattered occurrences south of Auckland as far as Foxton.

Hyridella menziesi menziesi (Gray, 1843). Widely spread throughout New Zealand, with many ecotypes.

Hyridella menziesi depauperata (Hutton, 1883). Lakes north of Auckland.

When compared with Suter's arrangement of four species and seven subspecies the above classification may appear somewhat iconoclastic. Of the forms admitted by Suter and here dismissed, lessoni, rugulosus, “aucklandicus” and lucasi can never have much standing. Unless hochstetteri of Dunker is a very

Picture icon

Fig. 1.—Hyridella aucklandica websteri (Simpson). Kaitoke Lake, Wanganui 58.4 × 30.4 mm.
Fig. 2.—Hyridella aucklandica aucklandica (Gray). Karlkohe. 67.8 × 32.2 mm.
Fig. 3.—”Hyridella aucklandica websteri (Simpson) Cotype ex Suter Collection 59.4 × 31.8 mm.
Fig. 4.—”Hyridella lessoni Kuster” Lake Wakatipu 50.0 × 29.0 mm.
Figs. 5, 6.—Hyridella aucklandica aucklandica (Gray). Type. Photographis by courtesy of British Museum (Natural History).
Fig. 7.—”Diplodon Menziesi Lucasi Suter.” Holotype, Lake Manapouri 44.3 × 24.0 mm.
Fig. 8.—Hyridella aucklandica qucklandica (Gray). Kaeo. 93.2 × 44.2 mm.
Fig. 9.—Hyridella menziesi depauperatus (Hutton). (Topotype) Lake Takapuna 68.2 × 39.9 mm.
Fig. 10.—Hyridella menziesi depauperatus (Hutton). Lectotype. Lake Takapuna. 58.3 × 30.7 mm.

Picture icon

Fig. 1.—”Diplodon Menziesi acutus Suter.” Holotype. 69.0 × 38.2 mm.
Figs. 2, 7—”Unio Menziesi Gray” Type. Photographs by courtesy of British Museum (Natural History).
Fig. 3.—Hyridella menziesi menziesi (Gray). Lake Rotorua 62.4 × 47.3 mm “hochstetteri” form.
Fig. 4.—Hyridella menziesi menziesi (Gray). Ohinemutu. Lake Rotorua 74.1 × 51.2 mm.
Fig. 5.—Hyridella menziesi menziesi (Gray). Motuopa, Lake Taupo 37.3 × 39.7 mm. (Topotype.)
Fig. 6.—”Unto zelebori. Dunket” Type (?) after Frauenfeld.
Fig. 8.—Hyridella menziesi depauperatus (Hutton). Lake Omapere 69.1 × 43.7 mm Topotype of “acutus, Suter”.

Picture icon

Fig. 1—Hyridella aucklandica websteri (Simpson). Lake Kaitoke, Wanganui.
Figs 2, 3—Hyridella aucklandica aucklandica (Gray). Type. Photographs by courtesy of British Museum (Natural History).
Fig. 4.—Hyridella aucklandica aucklandica (Gray). Kaeo.
Fig. 5, 6.—Hyridella menziesi menziesi (Gray). Type. Photographs by courtesy of British Museum (Natural History).
Fig. 7.—“Unio lutulentus Gould.” Cotype e [ unclear: ] Suter Collection. 60.3 × 29.3 mm.
Fig. 8.—Hyridella aucklandila aucklandica (Gray). Kaeo.

– 237 –

different shell from hochstetteri of Suter et al separation from menziesi will never prove to be easy, if possible. The three doubtful forms are lutulentus, and acutus, which if accepted by other workers should be classified as subspecies of menziesi, and zelebori, which may later prove to be a separate species. In addition there is a very wide form from Kaihoka Lake, West Wanganui Inlet, and a shell from Otago streams which may be separable.

The only published study of the life history of any New Zealand fresh water mussel is that of Perclval (1931). As regards distribution, the areas from which Hyridella has not been collected are of some significance. Specimens have been obtained from Spirits Bay and from Awanui. They have not been recorded from the consolidated sand dune country between. No records are known to the writer from Great Barrier Island or other off shore islands. In the South Island the genus has not been recorded from the Fiordland area nor from Stewart Island. Nor is it known from the Chatham Islands nor from the Subantarctic Islands.

Bibliography

Agnell, I., 1949. The Shell Morphology of some Swedish Unionides as Affected by Ecological Conditions. Ark. f. Zool., Bd. 41, hfte. 4, no. 15.

Ball, G. H., 1922. Variations in Fresh-water Mussels. Ecology, vol. 3, pp. 93-121.

Cotton, B. C. and Gabriel, C. J., 1932. Australian Unionidae. Proc. Roy. Soc. Vict., vol. 44 (n.s.) (2), pp. 153-160.

Cotton, B. C. and Godfrey, F. K., 1938. The Molluscs of South Australia, Pt. 1. The Pelecypoda. Govt. Printer, Adelaide, 314 pp.

Eagar, R. M. C., 1948. Variations in Shape of Shell with Respect to Ecological Station. A Review Dealing with Recent Unionidae and Certain Species of the Anthracosiidae in Upper Carboniferous Times. Proc. Roy. Soc. Edinb, vol. 63, pp. 130-148.

Iredale, T., 1934. The Fresh-Water Mussels of Australia. Austral. Zool., vol. 8 (1); pp. 57-78.

Mayr, E., 1942. Systematics and the Origin of Species. New York.

Ortmann, O. E., 1920. Correlation of Shape and Station in Fresh-Water Mussels. Proc. Amer. Phil. Soc., vol. 59, p. 269.

Percival, E., 1931. A Note on the Life History of Diplodon lutulentus Gould. Trans. N.Z. Inst., vol. 62, pp. 86-91.

Suter, H., 1905. Report on the Mollusca collected by Messrs. Keith Lucas and G. L. Hodgkin in Six Lakes of New Zealand. Trans. N.Z. Inst., 37, pp. 233-257.

— 1907. Descriptions of New Non-Marine Shells from New Zealand. Proc. Mal. Soc., vol. 7, pp. 236-240.

— 1913. Manual of New Zealand Mollusca. Govt. Printer, Wellington, N.Z.