Other formats

    TEI XML file   ePub eBook file  


    mail icontwitter iconBlogspot iconrss icon

Tuatara: Volume 10, Issue 1, April 1962

New Zealand Marine Provinces — Do They Exist?

New Zealand Marine Provinces — Do They Exist?

The marine biological provinces used in one form or another by most workers in New Zealand were originally based upon the distribution of mollusca. At the present time, when there is renewed interest in the distribution of marine animals, attempts are being made to adapt the molluscan provinces to fit other groups, without full realisation that the bases for such provinces have never been fully analysed. In a recent issue of this journal (Tuatara 9 (1), 1961 Powell has traced the development of the provinces and commented upon characteristic forms for each province.

The writer believes that if these provinces have any precise boundaries this has never been adequately demonstrated. It seems proper to put forward the following ideas at this time to show that not all workers on New Zealand mollusca accept the classical provincial divisions. These personal ideas have been developed over some fifteen years, while an immense quantity of distributional records have been accumulated. These records have not yet been finally analysed but the general trends crystallising from this study are here presented as a basis for discussion and criticism. Many of the distributional records have not yet been published but the writer's ideas have been freely expressed in discussion.

page 44

Before discussing distribution patterns themselves there are several general points which must be clarified. First and foremost, if we are going to use provinces we should be quite clear in our minds as to what we envisage as a province, what methods we should use in defining them, what is likely to happen at boundaries between provinces, and perhaps what use they can be expected to serve when we have completed what will certainly be a laborious analysis. Division of the faunal regions of the world into provinces must be based upon the belief, whether it is expressed or not, that within a region more than one set of fairly sharply differentiated physical, hydrological or historical factors must operate, or have operated, and that a majority of organisms of various groups must be similarly affected by such factors. The geographical ranges of these organisms should be so governed by such factors that the end-points of all their ranges will fall about the same point or within a relatively restricted area. This faunal boundary should be correlated with a physical, a hydrological, a historical or some other boundary. If such boundaries cannot be shown to exist, it seems doubtful if the provinces themselves exist.

Mere physical distance does not seem sufficient grounds for subdivision, providing there is continuity of habitat and an elongated coastline showing little change other than a gradual temperature gradient will show only a gradual faunal change providing no other major factors come into play. The enormous area covered by the Indo-Pacific shallow water marine fauna is sufficient evidence that distance in itself is not sufficient to create major marine provinces.

In a number of cases where detailed analyses of the distribution of organisms has been undertaken within regions it has proved exceedingly difficult to find or define boundaries acceptable to all workers. On the West Coast of North America two separate attempts to analyse the distribution of mollusca gave two widely variant results. Schenck and Keen (1936) analysed the distribution of the mid-points of ranges for some 1.948 molluscs as a basis for subdivision. The obvious areas where a large number of mid-points occurred were taken to represent the centres of the faunal provinces. Newell (1948) analysed the same data but used rate of change of fauna as the method for detecting faunal breaks and produced a very different result. The boundary areas on the two different schemes were:
Schenck And KeenNewell
62° - 58° N.72° N.
56° N.
48° -42° N.34½° N.
23° N.23° N.

Considering the number of writers who have discussed provinces, the number of definitions of a province are surprisingly few, and page 45 those few are very general. Woodward (1856, p. 349) in a pioneer subdivision of the world into molluscan provinces stated: ‘ In order to constitute a distinct province it is considered necessary that at least one-half the species should be peculiar.’ Schenck and Keen (1936, p. 923) have two statements which clarify their concept of a province. ‘ A marine molluscan province is a subdivision of a region, sub-region, or realm, populated by a distinctive assemblage of species; and, ‘A province has at its center a high percentage of species with restricted ranges: the periphery on the contrary, is marked by species of wide distribution. The size of a province being variable, its limits are best defined by statistical means.’ Newell disagreed violently with this latter concept and showed that the provinces outlined by Schenck and Keen were at variance with those used by other workers. Powell (1961) has recently stated,

‘A province may be defined as an area within a faunal region that exhibits a marked percentage of endemism.’ The concept of a province has been used in a very different sense to the above. For example Schmidt (1954, p. 328) in a general consideration of faunal realism, regions and provinces used the term ‘province’ for a very much wider biogeographic unit, his New Zealandian Province being contrasted with Oceanic and Antarctic provinces.

It is obvious from a consideration of the above definitions that apart from Woodward no worker has put forward objective criteria by which provinces may be recognised. The criteria accepted by other workers must therefore be decided from a consideration of their actions.

Powell (1961) has traced the historical development of the New Zealand molluscan provinces but it must be stressed that the provincial boundaries have changed considerably since Finlay in 1925 first formulated the basic pattern. The changes especially in the southern boundary of the Aupourian which are well described in Powell's paper are of such magnitude that the original evidence for each province is no longer applicable. The scheme currently accepted by Powell cannot in fact be based upon any recent analysis of molluscan distribution. Indeed information published on the distribution of marine mollusca to date is remarkably sketchy. Of the local checklists available only two cover the complete areas of provinces: that for the Antipodean (= Rossian) (Powell, 1955) and that for the Moriorian (Dell. 1960). For the rest of the New Zealand region the checklists available are:—


Powell (1940) published a list of the mollusca of the Aupourian Province, the boundaries then accepted being Ahipara on the west coast and Whangaroa on the east coast. Many additional records have been published but these in toto cover only the northern extremity of the province as now defined.

page 46


No lists are available.


Powell (1939) published a list of the mollusca of Stewart Island. There have been many subsequent additions, most of which have been published. Fleming (1950) presented a list of the mollusca of the fiords of western Southland based on material from Preservation Inlet, and Chalky, Dusky, Breaksea, Daggs. Doubtful and George Sounds, plus earlier collected material from Milford Sound to Puysegur Point. Numerous species have since been collected. Powell (1955) recorded all the species known from the Snares. These three lists combined do not cover completely the area usually incorporated in the Forsterian Province.

The actual length of the mainland coast covered even by lists of this type and vintage is thus comparatively small and no faunal records of this sort are available for any of the mainland boundary areas. The extremities are well documented and the differences are so much the more marked. Such lists as are available are complete lists of all mollusca known regardless of depth or habitat. Thus intertidal shelf and deep water forms are listed alongside pelagic species and are included in most analyses and comparisons based upon them. Any division into provinces must therefore be largely arbitrary, based upon individual knowledge of ranges and individual assessment of various forms as the most valuable ‘characteristic’ species. The steadily southward movement of the boundary between the Aupourian and the Cookian and the change in the list of species considered characteristic of the Aupourian represent therefore changes of opinion as well as increased knowledge.

The difficulties involved in listing typical forms for the various provinces may be demonstrated by considering some of the examples recently quoted by Powell (1961, pp. 4-8). In discussing the Aupourian Powell lists a number of East Australian forms which occur in the north of New Zealand. Quite a large number of such forms occur in the north but individual species have very different southern boundaries and some have very restricted ranges. It hardly seems justified to claim as typical Aupourian forms species which occur only in the extreme north of the Northland peninsula. Others range much more widely, such a species as Xenophalium pyrum reaching as far south as Kaikoura while the crab Lyreidus occurs in Cook Strait. Of the ‘characteristic elements: of the Aupourian, Xenophora neozelanica and Tonna haurakiensis range at least as far south as Wanganui on the west coast. Gomphina, Alcithoe depressa page 47 and Venericardia reinga have very restricted ranges in the far north of the North Island, while the heart urchin Brissus gigas and the brachiopod Terebratella haurakiensis as well as Tonna and Xenophora are shelf species. The distribution of shelf species will be discussed briefly later but patterns on the shelf do not appear to be similar to the distribution patterns of intertidal and shallow water species.

Four species only are quoted as ‘characteristic Cookian stenozonals’. Of these Penion ormesi is a shelf form and the distribution of Cellana denticulata will be discussed later.

There is therefore relatively little evidence for the existence of characteristic species of Aupourian or Cookian Provinces or for species whose ranges terminate at, or near, East Cape, the boundary between the Aupourian and the Cookian.

One of the dangers inherent in the use of provinces is that a species tends to become fixed in a worker's mind, as typical of the province from which it is first described. Subsequent discovery in another province is then often interpreted as demonstrating faunal influences. Thus Powell (1961, p. 4) cites Fusitriton laudandus as a Forsterian form occurring in the Aupourian, refers to northern occurrences of Cellana denticulata as ‘extra limital’, and again (p. 5) discusses the Cookian occurrences of warm-water genera such as Galeodea triganceae and Ranella multinodosa and (p. 6) interprets the occurrence of Neothyris lenticularis off East Cape as a possible survival of anotherwise restricted Forsterian element.

The writer has recently shown (Dell, 1956) that Fusitriton laudandus and Galeodea triganceae are typical archibenthal species wherever the archibenthal has been adequately sampled and it is very probable that the occurrence of odd shells of Fusitriton in the north is due to invasion of the shelf from the archibenthal. The occurrance of Galeodea on the shelf south of Banks Peninsula is mirrored by the occurrence of many such archibenthal forms on the shelf in this area.

Cellana denticulata is a common limpet north of Kaikoura and although sporadic in occurrence north of East Cape, it may be locally quite common as far north as the Three Kings. If full analysis had preceded the creation of provinces, this species would be treated as typical of the east coast from Kaikoura to East Cape and on headlands and off-shore islands through the east coast of the North Island. Similarly Neothyris lenticularis, which is also quite common in Cook Strait, would be considered to range from Stewart Island to East Cape and not to be a typical Forsterian form. In like fashion Ranella would be accepted as being just as typical of the Otago shelf as it is of the east Wellington shelf and the Bay of Plenty.

Mention has been made of the fact that distribution patterns of shelf species are rather different from those shown by intertidal and page 48 shallow water forms. A few examples are given below for some characteristic shelf genera, mainly for the east coast from Coromandel to the Otago Peninsula since so much detailed evidence is now available for this area.

The large siphon-whelk genus Penion has some fifteen species recorded from New Zealand waters. A complex of forms occurs in the Bay of Plenty but P. adusta is undoubtedly the dominant shelf species in one form or another extending as far south as Hawrkes Bay. From Castlepoint to Banks Peninsula adusta is replaced by P. ormesi while from Banks Peninsula to Foveaux Strait P. fairfieldae is represented. Other species are found in the area under consideration. Thus P. dispar occurs off Cape Campbell and odd specimens of P. benthicola are known from the Otago shelf, but neither species is ecologically important. The helmet shell genus Xenophalium is well represented on the shelf by some eleven species. Again a number of species occur in the Bay of Plenty but the dominant form in the north is X. pyrum which reaches as far south as Kaikoura. From Castlepoint to Cape Palliser X. abernethyi largely replaces pyrum while X. hamiltoni occurs with abernethyi off Castlepoint but becomes the dominant species from Cape Campbell to Kaikoura. From Banks Peninsula to the Otago Peninsula X. finlayi seems the sole representative.

The endemic volute genus Alcithoe occurs on the New Zealand shelf in bewildering variety. Some populations would seem to defy systematic analysis but some species can be identified more readily. Alcithoe jaculoides apparently occurs in two forms in the Bay of Plenty and one of the forms is not uncommon south of Banks Peninsula but has not been recorded between. A. larochei is distributed in the Bay of Plenty, has been recorded from off Castlepoint and is one of the two dominant forms off Cape Campbell and Kaikoura. Alcithoe calva is known from off Castlepoint, and is the other dominant off Cape Campbell. South of Banks Peninsula a variant of A. calva is one of the commonest forms intermingled with typical calva.

Four species of Aeneator are recorded in the area under consideration. Aeneator compta and A. marshalli separabilis both occur in the Bay of Plenty but neither have been recorded south of Fast Cape. Two subspecies of Aeneator otagoensis range from Hawke Bay to the Otago Peninsula, A. o. cookiana extending from Hawke Bay to Kaikoura and A. o. otagoensis recorded from Timaru to the Otago Peninsula.

While East Cape forms a southern boundary for some of these forms. Castlepoint and Banks Peninsula seem equally valid boundaries for others. At the same time other common shelf species such as Austrofusus glans range from north to south in the area under consideration unchanged.

page 49

What little evidence is available tends to show that the archibenthal fauna is much more uniform and widely distributed. Pelagic forms on the other hand are much more clearly tied to the surface water distribution pattern and many of them fluctuate north and south with seasonal fluctuations of the main surface water masses. Such forms can hardly be considered in a discussion of provinces.

On the criteria accepted in practice by other workers it is perhaps surprising that not more provinces have been erected. Taking the actions of workers as implicit criteria for provincial separation and analysing the number of endemic forms in the faunal lists presented, the figures for the various areas are as follows:
Aupourian (in Powell's 1940 sense)64926140%
Forsterian (Stewart Island)3628022%
Moriorian (Dell, 1960)3204915%
Antipodean (Powell, 1955)3668824%
Macquarie Island503166%

It is surprising on these figures that Macquarie Island with such a high percentage of endemic forms should be the one area which is not considered to be a province in its own right but is included in the Kerguelenian together with Kerguelen and Heard Islands. It would appear from the above that a percentage of endemic species as low as 15% or 22% or 24% is considered sufficient to designate a province. But the Auckland Islands have 16% of endemic forms, the Snares 18% and the Bounties 23%. To be consistent these areas should also become separate provinces.

In the case of some areas factors other than percentage of endemism have been considered. In discussing the Moriorian (Chatham Island Province) the lack of some common mainland forms has also been taken into account. Whether the lack of species of certain groups is a valid criterion for separation is an arguable point.

More Dynamic Approach Required

My own view, based largely on knowledge of the distribution of mollusca. is that no clear boundaries can be defined for any of the mainland littoral provinces, that shelf distribution patterns do not by any means coincide with those for littoral forms and that the concept of provinces as regards the mainland of New Zealand has largely outlived its usefulness. A more dynamic approach would envisage the New Zealand mollusca as being composed very largely page 50 of three groups, one extending essentially throughout New Zealand, a second essentially northern in origin, members of which extend to a varying degree to the south, and a third group essentially southern in origin, the members of which extend to a varying degree to the north. There is no thought that these three groups are at all homogeneous in composition and much work remains to be done to analyse them completely. It is, however, firmly believed that energy is better spent in compiling full distribution data for all our species than in attempting to define rather nebulous provincial boundaries.

This position does not appear to be at variance with the practice of other workers. Knox for example states (1960, p. 618), ‘It is evident there are two main centres of distribution, a warm-water one in the north and a cold-water one in the south. From each of these centres, which are characterized by endemic species and non-endemic species confined to them, numbers of species extend varying degrees north and south, forming an extensive transition zone which is related to the fluctuations in hydrological conditions of the mixed waters along the central east coast.’

Similar views have been expressed as regards the distribution of New Zealand fishes by Moreland in 1959 and in a paper read at a meeting of the New Zealand Marine Sciences Society in 1961. The distribution of marine algae (Moore, 1961) and echinoderms (Pawson, 1961) could undoubtedly be discussed equally as well within such a framework as within provincial compartments.

There seems little doubt that there is a well defined faunal break between the southern islands of New Zealand (Campbell, the Aucklands, Bounties and Antipodes) and the New Zealand mainland including the Snares. At the same time many of the really characteristic Antipodean genera and species cross this faunal break and extend to a varying degree to the north. Chlamys (Zygochlamys) delicatulus for example occurs off Macquarie, Campbell, the Aucklands, Antipodes and extends up the east coast of the South Island probably as far as Cook Strait. The very characteristic Subantarctic genus Gaimardia has one species on Macquarie, two in the southern islands and one in the southern part of the South Island. Kidderia, which also appears Subantarctic in origin, has three species at Macquarie, two in the southern islands and five on the New Zealand mainland. The genus Margarella which is widely represented in the Antarctic and Subantarctic has one species on Macquarie, two in the southern islands, one on the Chathams and three on Stewart Island and the southern part of the South Island. One of these latter, M. rosea, extends as far north as Kaikoura. Limpets of the Cellana strigilis group have one species on the Aucklands and Campbell, another on the Antipodes, one on the Bounties, one on the Snares and one on the Chathams while one page 51 C. s. redimiculum is recorded from Stewart Island to Kaikoura. Macquariella has one species on Macquarie, and one at the Antipodes while one has been recorded from the Aucklands, the Chathams and Stewart Island. Kerguelenella has one species on Macquarie, one on Campbell, the Aucklands and Antipodes, and another at the Snares and Stewart Island.

The above examples are taken from amongst the most characteristic elements of the mollusca fauna of the southern islands and show the varying degree to which this southern influence extends to the north.

Nerita and Saxostrea are two intertidal northern forms which show the extent to which such species can extend southwards. Saxostrea glomerata, the common Auckland rock oyster, is a marked feature of rocky coasts, forming obvious intertidal bands as far south as the Bay of Plenty. Towards East Cape artificially transported colonies have survived but no spat settles yet Occasional living specimens have been collected as far south as Cook Strait. Nerita melanotragus is an extremely common northern intertidal species, usually very obvious where it occurs. It becomes less common in the southern part of the North Island and in Cook Strait is very local and rare. Occasional specimens occur on Stephen Island, d'Urville Island and at Nelson.

Enough illustrations have probably been given to indicate the trends of distribution patterns. Evidence that more widespread fluctuations occurred, especially northern extensions of the cold-water southern fauna in Waitotaran and Nukumaruan times and possible southward extensions of the warm-water northern fauna in the Castlecliffian, has been detailed by Fleming (1944).

This discussion is presented at this time in the hope that it may stimulate workers in other groups to examine the concept of provinces in New Zealand and to publish complete data on geographical ranges of individual species. The writer is continuing to record data on the distribution of marine mollusca, which it is hoped will eventually be analysed and published.


Dell. R. K., 1956. The Archibenthal Mollusca of New Zealand. Dom. Mus. Bull. 15.

—— 1960. Chatham Island Marine Mollusca based upon the Collections of the Chatham Islands Expedition, 1954. N.Z.D.S.I.R. Bull. 139; 141-57.

Fleming, C. A., 1944. Molluscan Evidence of Pliocene Climatic Change in New Zealand. Trans. Roy. Soc. N.Z., 74; 207-20.

—— 1950. The Molluscan Fauna of the Fiords of Western Southland. N.Z. Journ. Sci. Tech. B., 31 (5); 20-40.

Knox, G.A., 1960. Littoral ecology and biogeography of the southern oceans. Proc. Roy. Soc. B., 152; 577-624.

Moore, L. B., 1961. Distribution Patterns in New Zealand Seaweeds. Tuatara, 9: 18-23.

Moreland, J., 1959. The Composition. Distribution and Origin of the New Zealand Fish Fauna. Proc. N.Z. Ecolog. Soc, 6: 28-30.

Newell, I.M., 1948. Marine Molluscan Provinces of Western North America: A Critique and a New Analysis. Proc. Am. Phil. Soc. 92; 155-66.

Pawson, D.L., 1961. Distribution Patterns of New Zealand Echinoderms. Tuatara, 9: 9-18.

Powell, A. W. B., 1939. The Mollusca of Stewart Island. Rec. Auck. Inst. Mus., 2: 211-38.

—— 1940. The Marine Mollusca of the Aupourian Province, New Zealand, Trans. Roy. Soc. N.Z., 70; 205-48.

—— 1955. Mollusca of the Southern Islands of New Zealand. N.Z.D.S.I.R. Cape Exped. Bull. 15.

—— 1961. New Zealand Biotic Provinces. Tuatara, 9; 1-8.

Schenck, H.G., and Keen, A.M., 1936. Marine Molluscan Provinces of Western North America. Proc. Am. Phil. Soc., 76; 921-38.

Schmidt. K.P., 1954. Faunal realms, regions and provinces. Quart, Rev. Biol., 29; 322-31.

Woodward, S. P., 1851-6. A Manual of the Mollusca, London.