The Land Mollusca of Fiordland, South-West Otago

Transactions and Proceedings of the Royal Society of New Zealand, Volume 82, 1954-55, Page 1135

The Land Mollusca of Fiordland, South-West Otago R. K. Dell [Read before Wellington Brunch April 28, 1954; received by Editor, April 29, 1954.] Abstract The land snail fauna of the Fiordland area la listed. Thirty-three species and subspecies are recorded. Five new species, and three new subspecies are described. The origin and relationships of this fauna are discussed. Introduction Our knowledge of the land mollusca of the Fiordland area is due solely to Suter's records of the results of two small collections. The first made by Henry on Resolution Island before the turn of the century, resulted in the description of Phelussa costata (Suter) and P. henryi (Suter). A private scientific party visited the Sounds area in the Rakiura in 1908, and as a result of what must have been rather sporadic collecting, Suter (1913) records the following species: Phelusca helmsi (Hutton), Allodiscus dimorphus (Pfeiffer), Flammulina zebra (Le Guillou), Ptychodon monoplax (Suter), Fectola colensoi (Suter) and Fectola tapirina (Hutton). Of these Allodiscus dimorphus and Fectola colensoi were not known from any other South Island locality. The specimen of Fectola colensoi is in the Suter Collection and proves to be Fectola tapirina (Hutton). The Milford specimen of Allodiscus dimorphus cannot be traced but a species of Allodiscus allied to dimorphus, though very distinct, is described in this paper. During the New Zealand-American Fiordland Expedition, Mr. R. R. Forster, of the Canterbury Museum, and the writer collected land snails in the Caswell Sound-Stillwater River area, and Mr. P. C. Bull collected leaf mould from Resolution Island and Dusky Sound. The writer has already described Maoriconcha fiordlandica from Caswell Sound (Dell, 1952 A) and recorded Obanella allanae from Resolution Island and Dusky Sound (Dell, 1952 B). During January, 1953, Mr. R. R. Forster collected in the vicinity of Lake Te Au, and in February, 1953, Mr. G. W. Ramsay collected snails and leafmould from a number of Fiordland localities. Professor B. Percival and Mr. J. H. Sorensen have both collected material from the vicinity of Takahe Valley. The writer wishes to acknowledge his indebtedness to all these collectors for the use of material. The area considered in this paper is the western watershed and the coastal strip from Milford Sound to Puysegur Point, westward from Lake Te Anau. The area has been arbitrarily delineated in the north by a line from the mouth of the Clinton River in the north of Lake Te Anau to the headwaters of the Cleddau River, and in the south by a line from the South Arm of Lake Manapouri through the Cameron Mountains to Puysegur Point. This area may prove to be a land snail faunal entity although the actual eastern boundary cannot as yet, be clearly stated. There does appear to be a definite faunal break between the Fiordland area and the area stretching from Lake Wakatipu in the north, to the Longwood Range in the south. This latter area is characterised by the presence of Aeschrodomus of the stipulata type, which is completely lacking in Fiordland, as defined

above. More accurate delineation of these areas must await complete faunal analysis. Systematics Murdochia chiltoni (Suter) Resolution Island and Dusky Sound, P. C. Bull, —/3/1949. This species occurs in the southern part of the South Island and in Stewart Island. Murdochia cf. calvum (Hutton) Murdochia calvum (Hutton) is known only from a single shell collected by Helms at Greymouth. Suter redescribed and refigured the unique type, but his description and figure are not highly diagnostic. Unfortunately the type cannot at present be located in the Canterbury Museum, and no topotypes appear to have been collected. A moderately common shell in Fiordland appears to be related to calvum but until these shells can be critically compared with undoubted specimens of calvum, the identification cannot be certain. Localities. Lake Hankinson, Middle Fiord, Lake Te Auau, G. W. Ramsay, 14/2/1953; Stillwater River, Caswell Sound, R. K. D, 13/3/1949; Resolution Island and Dusky Sound, P. C. Bull, —/3/1949; Key Summit, Homer, W. R. B. Oliver, 26/12/1944. Phelussa henryi (Suter) Resolution Island (type). Very few specimens have been collected. Four shells have been seen, the type, one other specimen in the Suter Collection from Resolution Island, one from the same locality in the Canterbury Museum, and a specimen in the O'Connor Collection with no data. Phelussa costata (Suter). Resolution Island (type). The type series consists of three damaged shells. They are fairly constant as regards width of umbilicus and number of riblets. One specimen in the Canterbury Museum from the same locality agrees with the type series. Four shells, one from the Nuggets, one from Solander Island, and two in the Suter Collection labelled “West Coast Sounds” agree with typical costata in the number of riblets but have a wider umbilicus (comparable with that seen in henryi). For ease of reference shells with a wide umbilicus have been called Phelussa costata Suter, Type B. Any adequate discussion of the problem is hampered by the lack of significant series and the lack of locality data. The types of both P. costata and P. henryi come from the same locality, and it is quite likely that more specimens would show that these are but two extremes of one variable species. A combination of the ranges for the three forms, P. henryi, P costata and P. costata Type B, and combined means do not show an extreme degree of variation. * The abbreviations used in this paper for measurements and indices are the same as those used in a previous paper (Dell. 1954)—i.e., D. is diameter in mm.; H. is height in mm.; S, is height of spire in mm; U, is diameter of umbilicus in mm.; R. is the number of riblets on the body whorl; H.I, is the Height Index, the height expressed as a percentage of the diameter; S.I, is the Spire Index, the height of the spire expressed as a percentage of the total height; U.I., is the Umbilical Index. the diameter of the umbilicus expressed as a percentage of the total diameter; and R. I, is the Riblet Index, the number of riblets on the body whorl expressed as a percentage of the diameter in mm. H.I., 50–63 (56); SI, 19–38 (31); Ul., 4–17 (12), RI, 7–15 (10).

Ranges and Means of Indices for Specimens of the henryi-costata complex H.I. S.I. U.I. R.I. P. henryi 50–57 (53) 23–38 (30) 15–17 (16) 7–9 (S) P. costata Type B 57–63 (60) 19–35 (29) 11–16 (13) 12–15 (13) P. costata 54 32 4–6 (5) 10–12 (11) Phelussa helmsi (Hutton) Localities. Head of Caswell Sound; Stillwater River and Leslie Clearing, R. K. D., –/3/1949; Lake Te Au, near South Arm of Lake Te Anau, R. R. Forster, —/1/53; Wapiti River, Western Te Anau, G. W. Ramsay, 16/2/1953; Treble Mountain, Preservation Inlet, 2,500 feet, O 'Connor Collection. Outside the Fiordland area, this species occurs throughout a large part, of the southern South Island. Allodiscus yenulatus Pfeiffer. A thin fragile shell apparently closely allied to, and possibly identical with, Allodiscus venulatus Pfr occurs throughout Fiordland. The shell appears to be constantly smaller than specimens labelled venulatus in the Suter Collection, the ribs are slightly more numerous and the shell is more fragile. Until Pfeiffer's species can be exactly determined and until the range of variation of venulatus is more fully known it does not seem advisable to differentiate the fiordland form. Localities. Bowen Palls, Milford Sound, J. T. Salmon, 19/12/1944; Stillwater River, Leslie Clearing and foreshore of Caswell Sound, R. K. D. and R. R. Forster, —/3/1949; Treble Mountain, Preservation Inlet, O'Connor Collection; Lake Te Au, near South Arm, Lake Te Anau, R. R. Forster, —/1/1953. A specimen from Caswell Sound has a diameter of 4.18 mm, 114 riblets on the body whorl and a Riblet Index of 27 A specimen in the Suter Collection from Capleston, Reefton has a diameter of 5.36 mm., 117 riblets on the body whorl and a Riblet Index of 22. Allodiscus planulatus (Hutton) Leslie Clearing, Stillwater River, and foreshore of Caswell Sound, R. K. D, —/3/1949. This species is widely spread throughout New Zealand, extending as far south as the Auckland Islands. Allodiscus anstrodimorphus n. sp. Text-fig. 1, Fig. 1. Shell of similar build to the dimorphus series but considerably smaller when adult (5 whorls). It appears to be closest to A. fallax Powell, but differs in the much smaller adult size, slightly fewer riblets and in having the spirals much stronger, reticulating the secondary radials to form small squares and fenestrating the primary radials. This latter feature allies it to A. spiritus Powell, from which it is readily distinguishable by the much sparser radials. Whorls five, including a protoconch of 1½ whorls. Protoconch at first microscopically pustulated, with incised spirals developing between the pustules over the last whorl. Post-nuclear sculpture consisting of well defined axial ribs, 62–72 on the penultimate, 79–102 on the body whorl (Riblet Index 17–18). Interstices with 7–9 secondary radials, somewhat variable in strength. Spiral sculpture comparatively strong, crossing the primary axials. Spire considerably more raised than in other members of the dimorphus series. Shell narrowly perforate when young, but perforation soon closed by the reflexed inner lip and columella and imperforate when adult.

Fig. 1. —Allodiscus austrodmorphus n.sp. Holotype. 5.18 × 3.45 mm. Figs. 2, 5.—Allodiscus fectolodets n.sp. Holotype. 4.27 × 2.18 mm Figs 3, 6, 10.—Flammulina lateaperta n.sp. Holotype. 4.09 × 2.27 mm. Fig. 4.—Ptychodon fiordlandica n.sp. Holotype. 1 19 × 1.04 mm. Fig. 7. —Ptyhodon gadus obscurus n.subsp. Holotype. 1.5 × 0.91 mm. Fig. 8.—Phrixgnathus viridula viridula Suter. Type 2.41 × 1.77 mm. Fig. 9.—Ptychodon blacki n.sp Holotype. 1.64 × 1.13 mm Fig. 11.—Phrixgnathus viridula caswelli n.subsp. Holotype 2 27 × 1.5 mm. Fig. 12. —Aperture of Ptychodon black n.sp. Holotype Fig. 13—Aperture of Ptychodon blacki n.sp. Leslie Clearing. Fig. 14.—Aperture of Ptychodon fiordlandica n.sp. Holotype. Fig. 15 —Teeth from radula of Flammulina lateaperta n.sp. R. K. Dell, del.

Holotype (M. 6149) and paratype in the Dominion Museum, paratypes in the Canterbury Museum. Localities. Stillwater River, Caswell Sound, R. K. D., 13/3/1949 (halo-type); Leslie Clearing, Caswell Sound, R. K. D., 15/3/1949; Lake Te Au, near South Arm, Lake Te Anau, R. R. Forster, —/1/1949. Measurements. D. H. S. I. H.I. S.I. R.I. Stillwater River (Holotype) 5.18 3.45 0.91 91 67 26 18 Leslie Clearing 5.64 3.64 0.91 102 64 25 18 Lake Te Au 4.36 2 73 0.54 79 63 20 18 Lake Te Au 5.45 3.54 0.91 91 65 26 17 The ranges and means of the indices are: Height Index, 63–67 (65); Spire Index. 20–26 (24); Umbilical Index, 18–18 (18). Allodiscus fectoloides n.sp. Figs. 2, 5. Shell small, discoidal, perforated, very finely ribbed. Whorls four, including a protoconch of 1½ whorls. Whorls rapidly increasing. Protoconch somewhat raised, sculptured with fine spirals. Post-nuclear sculpture consisting of fine radials which gradually become more oblique and flexuous, and retractive at the suture, some 167 on the body whorl of the holotype (Riblet Index 39). Interstices with very fine growth lines, reticulated with incised spirals which also cross the primary axials. Colour of upper surface pale corneous with irregular darker blotches, a regular series of lighter spots close to the suture. Base tessellated Spire flattened, slightly raised. Suture impressed. Aperture very wide, oblique. Peristome retracted at the suture forming a distinct sinus, advancing above the periphery. Umbilicus comparatively wide, deep, perspective. Localities. Lake Te Au, R. R. Forster, –/1/1953 (holotype); Lake Hankinson, west of Te Anau, G. W. Ramsay, 14/2/1953 (paratype). Holotype in Canterbury Museum, paratype in Dominion Museum (M. 6152). This species differs from all known species of Allodiscus in the rapidly expanding whorls and the sutural sinus. It agrees with Allodiscus in the details of the protoconch, the type of sculpture and the basal tessellation. Allodiscus fectoloides appears to be restricted to Fiordland, and so far as is known is found only in the eastern section D. H. U. R. H.I. U.I. R.I. Holotype 4.27 2.18 0.45 167 49 10 39 Paratype 3.82 1.91 0.45 143 50 12 37 Therasia decidua (Pfeiffer) Locality Notornis Valley, Middle Arm of Lake Te Anau, J. H. Sorensen, 11/1/1949. A widely distributed species throughout the rest of New Zealand, its peripheral occurrence in the Fiordland area would indicate that it is probably a recent addition to the fauna of this region. Thermia cressida (Hutton) Although recorded from Preservation Inlet by Hutton, the writer has seen specimens only from the Princess Range, Lake Poteriteri, collected by G. W. Ramsay. This latter area is outside the Fiordland area as here defined.

Flammulina zebra (Le Guillou). Dufky Sound, P. C. Bull, –/3/1949; Preservation Inlet (fide Suter). This widely spread species is often sporadic in occurrence. It will probably prove to be more widely distributed in the Fiordland area. Flammulina feredayi Suter. Locality. Head of Caswell Sound, R. K. Dell, 13/3/1949. This is another widely distributed species that penetrates into the Fiordland area. Flammulina lateaperta n.sp. Figs. 3, 6, 10, 15. Shell moderately large, very thin, yellowish-gold, imperforate. Whorls 2 ½, including a smooth protoconch of 1 ¼ whorls; rapidly increasing, body whorl greatly inflated. Sculpture consisting of rather regular growth wrinkles, most marked at the sutures, and one or two short spiral grooves, most marked at the aperture. By reflected light the rest of the surface appears smooth and shining, but by transmitted light fine irregular, rather disconnected spirals can be seen in the texture of the shell itself. Spire sunken, hardly raised above the level of the body whorl. Aperture wide and swollen. Columella thin, almost vertical. Lower lip of aperture drawn well back, exposing the inner shell whorls. There is no umbilicus but a narrow, shallow groove runs across the base of the shell parallel to the columella. Major diameter, 4.09 mm.; minimum diameter, 3.0 mm.; height, 2.27 mm. Locality. Stillwater River, Caswell Sound, R. K. Dell, 13/3/1949 (Holotype and paratypes); Lake Te Au, near South Arm of Lake Te Anau, R. R. Forster, —/1/1953 (paratypes), Puysegur Point, —/7/1940, O'Connor Collection. Holotype (M. 6153) and paratypes in collection of Dominion Museum, paratypes in Canterbury Museum. The distinguishing features are the rapid whorl inflation, the size, lack of umbilicus, the very fragile shell, the sculpture and the peculiar form of the aperture. The radula (Fig. 15) has the formula 18–8–1–8–18. Central and inner laterals similar in shape. Central tricuspid, with long slender mesocone and two weak lateral cusps. Bases on inner laterals becoming more oblique, entocone becoming larger in size and ectocone decreasing. About the 18th tooth entocone approximately the same strength as mesocone Ectocone then becomes obsolete. From about the 22nd tooth outwards the ectocone has disappeared and remaining teeth are bicuspid, with flat bases and oblique cusps. The animal is similar to that found in Flammoconcha and cannot be wholly retracted within the shell as also is the case with Flammulina feredayi Suter. This species has the appearance of a Flammulina in which the shell is beginning to degenerate and thus is on the morphological line of development between typical Flammulina and such forms as Maoriconcha and Otoconcha. The genus Flammulina as at present understood in New Zealand is probably polyphyletic. At least two different lines can be postulated, the first comprising the small shells, feredayi (Suter), pilsbryi Suter and costulata (Hutton) with spirally sculptured protoconchs and well marked radials and the rest with smooth protoconchs and less distinct sculpture. It is possible that Flammoconcha with its similar animal,

spiral protoconch and post-nuclear axials is derived from the first group and that Flammulina lateaperta represents a stage in the development of Maoriconcha and Otoconcha from the second group. Flammulina perdita (Hutton) Locality. Preservation Inlet, Suter Collection; Lake Hankinson, West of Lake Te Anau, G. W. Ramsay, 14/2/1953. Maoriconcha fiordlandica Dell. Localities. Head of Caswell Sound (type); Stillwater River and Leslie Clearing, Fiordland, R. K. Dell, –/3/1949; Wapiti River, West Te Anau, under stones, G. W. Ramsay, 16/2/1953. It would appear that Suter's record of Otoconcha dimidiata (Pfr.) from near Lake Te Anau really refers to this species as superficially (ho animals are very similar. Suter's actual specimen has not been located. Ptychodon gadus obscurus n.subsp. Text-fig. 1, Fig. 7. A small Ptychodon from Resolution Island and other localities in Dusky Sound is apparently a regional representative of the Codfish Island, Ptychodon gadus Dell. The Fiordland form differs in having a slightly wider umbilicus and much finer riblets (Riblet Index 41 to 46 as compared with 30 to 34 in P. gadus gadus). The ranges and means of indices for the two forms of gadus are:— H.I. U.I. R.I. Ptychodon gadus gadus Dell 60–64 (62) 18–19 (18) 41 46 (43) P. gadus obscurus n subsp 57–63 (60) 15–18 (16) 30–34 (32) D. H. U. R. H.I. U.I. R.I. Resolution Island (Holotype) 1.5 .91 .23 49 61 15 32 Resolution Island (Paratype) 1.45 — .27 50 — 18 34 Dusky Sound (Paratype) 1.59 .91 .27 53 57 17 33 Dusky Sound (Paratype) 1.5 .95 .23 45 63 15 30 Ptychodon gadus is the second case of a species occurring as a geographic subspecies in Fiordland and on Codfish Island. The other is Ptychodon monoplar known from Preservation Inlet (monoplax monoplax Suter) and Codfish (monoplax hiarara Dell). Holotype (M. 6154) and paratypes in Dominion Museum. Ptychodon monoplax Suter Suter described this species from Bisbee Bay, Preservation Inlet. It is characterized by a weak plait high up on the parietal wall. Dell (1954) described a subspecies (P. monoplax hiarara) from Codfish Island. off Stewart Island, which differs in the number of riblets and in the width of the umbilicus. A population from Notornis Valley, near the Middle Arm, Lake Te Anau has the riblets twice as numerous as monoplax monoplax Suter and is here separated as another subspecies. A few shells collected by Mr. R. R. Forster at Lake Te Au (near the South Arm of Lake Te Anau) are very similar to monoplax hiarara Dell. The sample is small and only one adult shell is present, but it does appear to differ from hiarara in having slightly more numerous riblets and a slightly narrower umbilicus. When more specimens from intermediate localities are examined it may be found that a cline is present, but at the moment the geographically separated populations are fairly constant and apparently discrete.

Ptychodon monoplax monoplax Suter. In the nominate form the spire is comparatively short (Spire Index 22 to 35, mean 30), the umbilicus narrow (Index 26 to 31, mean 29) and the riblets sparse (Index 34 to 37, mean 36). Locality. Bisbee Bay, Preservation Inlet. Holotype and paratypes in Suter Collection, now in the Dominion Museum. Ptychodon monoplax takahea n.subsp. This subspecies is characterized by the very numerous riblets (Index 65 to 69, mean 67). Locality. Notornis Valley, west of Middle Arm of Lake Te Anau, from leaf mould collected by J. H. Sorensen, 11/1/1949. Holotype (M. 6155) and two paratypes in Dominion Museum. D. H. S. U. R. H.I. S.I. U.I. R.I. Holotype 1.95 1.09 .32 .59 132 56 29 30 68 Paratype 1.91 1.13 .41 .59 125 59 36 31 65 Paratype 1.95 1.13 .41 59 135 58 36 30 69 Ptychodon monoplax cf. hiarara Dell. Three specimens from Lake Te Au are very similar to P. monoplax hiarara Dell but have slightly more numerous riblets and a slightly less open umbilicus. The sample contains only one adult shell, so these characters may not be constant. The ranges and means of the indices of the different forms of monoplax are:— H.I. S.I. U.I. R.I. monoplax monoplax 51–55 (53) 22–35 (30) 26–31 (29) 34–37 (36) monoplax hiarara 52–56 (54) 34–39 (36) 34–35 (35) 45–54 (49) monoplax lakahea 56–59 (58) 29–36 (34) 30–31 (30) 65–69 (67) Ptychodon blacki n.sp. Figs. 9, 12, 13 Shell very small, comparatively high, spire flattened, narrowly umbilicate. Whorls 4 ½, including a protoconch of 1½ slightly granular whorls Suture impressed, deep. Sculpture consisting of rather regularly spaced axial riblets, 103 on the body whorl of the holotype. These riblets become progressively more oblique and retractive at the suture. Interstices with fine growth lines crossed by fine spirals. Spire flattened, often obliquely flattened; upper whorls often at, or below the level of the penultimate whorl. Aperture lunate. Peristome retractive at the suture, forming a distinct open sinus, inflated on periphery. Apertural armature somewhat variable. One weak lamella above the middle of the parietal wall (sometimes absent, and usually set well back), a weak tooth forming a blunt angulation in the middle of the parietal wall, also set well back; two variably shaped, strong-based teeth on the columellar wall; and a delicate lamella high up on tie outer wall. From the columellar teeth a thickened band of white callus runs across the aperture, following the line of the axial riblets on the outer wall In some specimens this band is very well defined, and is divided by a series of numerous grooves into uneven, short lamellae. Columella very short, arcuate Umbilicus narrow and deep (Umbilical Index 11 to 19). Hole-type (M. 6156) and five paratypes in the Dominion Museum, a paratype in the Canterbury Museum.

Locality. Stillwater River, Caswell Sound, R. K. Dell, 13/3/1949 (Holotype); Leslie Clearing, Caswell Sound, R. K. Dell, 15/3/1949; Resolution Island. Dusky Sound, P. C. Bull, 6/3/1949. The arrangement of the teeth and lamellae is quite unlike that found in any other described species of Ptychodon. The Resolution Island specimens have a somewhat lower Riblet Index but the extremes almost overlap and the sample (two specimens) is small. Ptychodon blacki has a particularly high body whorl. Named for Mr. A. J. Black, the owner of the m.v. “Alert”, who so competently transported the New Zealand-American Fiordland Expedition. D. H. S. U. R. H.I. S.I. U.I. R.I. Holotype 1.64 1.13 .32 .27 103 75 28 17 62 Paratype 1.86 1.19 .32 .27 96 64 27 14 52 Paratype 1.68 1.0 .23 .32 103 60 23 19 61 Leslie Clearing 1.64 1.0 .18 .18 117 61 18 11 71 Leslie Clearing 1.59 .95 .23 .32 93 60 24 14 58 Resolution Island 1.54 1.0 .23 .23 69 65 23 15 45 Resolution Island 1.54 1.0 .27 .23 77 65 27 15 50 The means and ranges of indices for Ptychodon blacki are:— H.I. S.I. U.I. R.I. 60–75 (63) 18–28 (24) 11–19 (15) 45–71 (57) Ptychodon fiordlandica n.sp. Figs. 4, 14. Shell small, low, umbilicate. Whorls 4 ½, including a slightly granular protoconch of 1½ whorls. In young specimens the protoconch is sculptured with very fine raised riblets, but these are usually worn off. Sculpture consisting of regular raised axial riblets, 85 to 106 on the body whorl (94 on the holotype), which become somewhat retractive at the suture. Suture impressed. Spire slightly raised. Aperture semi-lunate, greatly reduced by the development of the teeth and the lamellae. Armature consisting of a very large bifid tooth on the middle of the parietal wall with two weak lamellae below, the lower almost obsolete and situated well back. Two strong teeth on the columella; one at the junction of the parietal wall and the columella with a very wide base, the tooth pointing horizontally across the aperture; the other, a very large tooth at the junction of the columella and the outer lip, curves upwards and outwards. Six evenly spaced, short but stout lamellae are spaced across the outer lip. Outer lip retracted at the suture to form a wide, shallow sinus. Umbilicus comparatively wide (Index 24 to 28), deep, perspective. Holotype (M. 6157) and many paratypes in the Dominion Museum. Locality. Lake Poteriteri, South-West Otago, from leafmould collected by G. W. Ramsay, 9/2/1953 (holotype); head of Caswell Sound, Fiordland, R. K. Dell, 23/3/1949. This species seems well characterized by the armature of the aperture, the lamellae of which are very well developed and very constant. The ranges of the indices show that the range of variation in all characters is comparatively slight. D. H. S. U. R. H.I. S.I. U.I. R.I. Holotype 1.91 1.04 .32 .5 94 54 31 26 49 Paratype 2.04 1.04 .32 .54 101 51 31 26 49 Paratype 2.0 104 .32 .54 106 52 31 27 53 Paratype 1.91 .95 .27 .54 86 49 28 28 45 Caswell Sound 1.82 .95 .27 .45 85 52 28 24 47

The ranges and means of indices for Ptychodon fiordlandica are:— H.I. S.I. U.I. R.I. 49–54 (52) 28–31 (30) 24–28 (26) 47–53 (49) Charopa bianca Hutton. Locality. Head of Caswell Sound, Stillwater River and Leslie Clearing, R. K. Dell,–/3/1949. Charopa anguicula (Reeve) Locality. Head of Caswell Sound, Stillwater River and Leslie Clearing, R. K. Dell,–/3/1949. Charopa (Pseudegestula) transenna brookesi Dell. Locality. Notornis Valley, Middle Arm, Lake Te Anau, from leafmould collected by J. H. Sorensen, 11/1/1949. Fectola sterkiana Suter Notornis Valley, Middle Arm, Lake Te Anau, from leafmould collected by J. H. Sorensen, 11/1/1949. This species occurs sporadically throughout the South Island. Fectola tapirina (Hutton) Locality. Stillwater River, Caswell Sound, R. K. Dell, 13/3/1949; Dusky Sound, P. C. Bull, —/3/1949: Preservation Inlet (fide Suter); Lake Te Au, South Arm of Lake Te Anau, R. R. Forster, —/1/1953; Milford Sound (Suter's record of F. colensoi Suter). Suter recorded Endodonta colensoi Suter from Milford Sound. This species has not been recorded from any other locality in the South Island. A specimen from Milford Sound in the Suter Collection labelled Endodonta colensoi proves to be inseparable from Fectola tapirina (Hutton). Suter (1913, p. 712) in a key to the group of species that are now included under Fectola used the degree of development of spiral sculpture or the lack of it as primary divisions. F. colensoi “with indistinct or obsolete spiral striae” is there differentiated from Fectola tapirina, “interstices without spirals”. Fectola tapirina (Hutton), including specimens labelled as such in the Suter Collection, has intersticial spirals so that this primary distinction immediately disappears. The relationships of F. tapirina to topotypic specimens of F. colensoi can only be determined when the genus is revised In any case the Milford Sound specimen is better referred to tapirina. Subfectola rakiura Powell. A single specimen from the Stillwater River, Caswell Sound, agrees very well with Stewart Island topotypes. The Fiordland shell has slightly more numerous riblets (152 on a shell 1.77 mm. in diameter compared with 138 on a shell 1.68 mm. in diameter) but too few specimens have been seen to determine the range of variability. Obanella allanae Dell. Locality. Resolution Island and elsewhere in Dusky Sound, P. C. Bull, —/3/1949. This species has been recorded elsewhere from Stewart Island and Governor's Bush, Mt. Cook.

Paralaoma sericata (Suter) Locality Head of Caswell Sound, Stillwater River and Leslie Clearing, R. K. Dell, —/3/1949; Resolution Island and Dusky Sound, P. C. Bull, –/3/1949. Phrixgnathus celia Hutton. Locality. Milford Sound, J. T. Salmon, 18/12/1941, Head of Caswell Sound. Stillwater River and Leslie Clearing, R. K. Dell, –/3/1949; Resolution Island, P. C. Bull, —/3/1949; Notornis Valley, Te Anau, J. II. Sorensen, 17/1/1949, Lake Hankinson, west of Lake Te Anau, G. W. Ramsay, 14/2/1953. Phrixgnathus regularis (Pfr.) Locality. Resolution Island and Dusky Sound, P. C. Bull, –/3/1949; Lake Hankinson, west of Lake Te Anau, G. W. Ramsay, 14/2/1953. Phrixgnathus viridula caswelli n.subsp. Text-fig. 1, Fig. 11. A form of Phrixgnathus in the area is allied to viridula from Capleston, but is more depressed and has much stronger axial and spiral sculpture, and a more depressed outline. For these reasons it is here separated as a subspecies of viridula under the above name. A figure of the type of viridula is given for comparison. Holotype (M. 6158) and 8 paratypes in the Dominion Museum. Locality Leslie Clearing, Caswell Sound, R. K. Dell, 15/3/1949; (Holotype): Stillwater River and Head of Caswell Sound, R. K. D., –/3/1949; Resolution Island, P. C. Bull, –/3/1949. D. H. S. U. H.I. S.I. R.I. Holotype 2.27 1.5 0.5 .18 66 33 8 Head of Caswell Sound 2.64 1.77 0.64 .18 67 36 7 Resolution Island 2.04 1.36 0.45 .18 66 33 9 P. viridula Suter (Holotype) 2.41 1.77 0.64 Nil 73 36 – (Paratype) 2.0 1.45 0.45 .18 72 31 9 The ranges and means for the indices of the two sub-species are:- H.I. S.I. U.I. P. viridula viridula 72–73 (72) 31–36 (33) 9 P. viridula caswelli 66–67 (66) 33–36 (34) 7–9 (8) Discussion From the Fiordland area as defined herein 33 species of land molluscs are now known. Murdochia chiltoni (Suter) Murdochia cf calvum (Hutton) Phelussa henryi (Suter) Phelussa costata (Suter) Phelussa helmsi (Hutton) Allodiscus venulatus (Pfeiffer) Allodiscus planulatus (Hutton) Allodiscus austrodimorphus n. sp. Allodiscus fectoloides n. sp. Therasia deciduas (Pfeiffer) Thermia cressida (Hutton)

Flammulina zebra (Le Guillou) Flammulina feredayi (Suter) Flamnulina lateaperta n.sp. Flammulina perdita (Hutton) Maoriconcha fiordlandica Dell Ptychodon gadus obscurus n.subsp. Ptychodon monoplax monoplax Suter Ptychodon monoplax takahea n.subsp. Ptychodon monoplax cf. hiarara Dell Ptychodon blacki n.sp. Ptychodon fiordlandica n.sp. Charopa bianca (Hutton) Charopa anguicula (Reeve) Charopa (Pseudegestula) transenna brookesi Dell Fectola sterkiana (Suter) Fectola tapirina (Hutton) Subfectola rakiura Powell Obanella allanae Dell Paralaoma sericata (Suter) Phrixgnathus celia (Hutton) Phrixgnathus regularis (Pfeiffer) Phrixgnathus viridula caswelli n.subsp. Of these forms, 11 (33%) are endemic to the area. This degree of endemism is comparatively high but is exactly comparable with the situation on Stewart Island, where the total number of species recorded is exactly the same (Dell, 1954). Of these endemic forms two (Ptychodon monoplax and Ptychodon gadus obscurus) have regional representatives on Codfish Island, off Stewart Island, one (Phrixgnathus viridula caswelli) has a regional subspecies on the west coast of the South Island, and one (Phelussa henryi) is related to a shell which occurs on Solander Island and on the Nuggets. The remainder of the endemic forms may be designated “strong” endemics—i.e., they are not obviously related closely to other species, at least on the specific level. Of the non-endemic forms, one (Subfectola rakiura) is known elsewhere only from Stewart Island. The other non-endemic forms are all comparatively widely spread at least in the southern part of the South Island. The drainage divide in Fiordland runs from Mt. Tutoka to the Cameron Mountains (Wellman and Willett, 1942, p. 293) and in the northern and southern extremities, this drainage divide appears to coincide with the faunal divide. In both areas the drainage divide comprises high ranges which extend well above the bush line and therefore form a very efficient barrier to the dispersal of terrestrial mollusca. In the centre of the region the shores of Lake Te Anau and Lake Monowai appear to mark the eastern faunal barrier. There are, however, lower level passes between Lakes Te Anau and Monowai and the West Coast. Even if the drainage divide in this central area has acted in general as a faunal barrier, routes for faunal migration have existed along these lower levels. The nothern limits of the area are unknown as no land snails appear to have been collected from the area immediately north of Milford Sound. Of all areas in New Zealand, the one under consideration must have been most radically affected during the Pleistocene glaciation. Willett (1950, p. 33, Fig. 6)

shows the whole of this area under glacial ice in the Pleistocene. Such a postulate would involve the total destruction of animal life in the area and subsequent invasion since the ice retreated. Even if small pockets remained free from ice they would belong climatically to the tundra or periglacial zone and few, if any of the moisture-demanding cryptozoic animals could survive there. If such conditions prevailed, any land molluscs present before the onset of glacial conditions must either have been exterminated or have migrated northwards before the advancing ice tongue. If conditions were uniformly as unfavourable for the animal life in leafmould as Willett has indicated, all the animals that survived must at least have reached what is now the North Island. With amelioration of glacial conditions in the south, reinvasions or new invasions from the northern stock of cryptozoic animals would take place. In areas such as Fiordland or Stewart Island, where the degree of endemism is comparatively high, it is difficult to believe that the degree of specific differentiation present could have taken place in the time available since the end of the Pleistocene glaciation. Such a course of events would explain the present sporadic distribution patterns of such land molluscs as Phenacohelix pilula (at Auckland and Stewart Island) and the genus Aeschrodomus (in the Waikato and in eastern Otago and Southland). It must be supposed in these cases that the species has subsequently died out in intermediate areas, or perhaps that the area surrounding the retreating glacier edges was the only suitable one for them, and that except in some few favoured localities, these forms migrated close to the edge of the ice edge but did not establish itself in intermediate localities. At present there is little evidence to show just what biological conditions existed in the southern part of the South Island during the glacial and interglacial periods of the major Pleistocene glaciation. Apparently there is some evidence for pockets of bush remaining on the West Coast of the South Island north of the Sounds, and for Olearia scrub in parts of the Fiordland area. Such areas would serve as refuges for some land molluscs. It is probable that a few of the Fiordland species were derived from West Coast forms—e.g., Phrixgnathus viridula caswelli, though in this case there has been subspecific differentiation. Of the non-endemic Fiordland forms, those that are widely spread elsewhere in the South Island may well represent invasions subsequent to the glaciation. The Stewart Island-Fiordland link could be explained in part if Stewart Island itself served as a refuge or a centre for re-invasion, but it is then difficult to explain why such forms do not occur in Southland or south-east Otago. Some of the well-marked endemic elements of the Fiordland land molluscan fauna may well have found refuge in pockets of scrub in the Fiordland area itself, if such existed throughout the period of glaciation. Such isolated pockets would form ideal localities in which speciation could proceed. The comparatively high number of forms of such a genus as Ptychodon could well be due to such a situation, coupled with subsequent invasions from possibly three outside areas. The present land snail fauna of the Fiordland area could therefore have been derived from the following sources:— (a) from forms which sheltered in scrub pockets within the area, during the Pleistocene glaciation; (b) from subsequent invasions from Stewart Island; (c) from subsequent invasions from the West Coast of the South Island;

(d) from subsequent invasions from the north, via the east coast of the South Island. There still remain a number of endemic forms which are not obviously related to other species whose presence in the Fiordland area cannot be adequately explained at present—e.g., Allodiscus fectoloides and Maoriconcha fiordlandica The internal distribution of land molluscs in the Fiordland area is not as yet adequately known. The peculiarly dissected topography should result in differential faunal distribution, but until more adequate collections are available this problem cannot be studied. Literature Cited Dell, R. K., 1952 A. Otoconcha and Its Allies in New Zealand. Dom. Mus. Rec. Zool, Vol. 2, pp. 59–69. – 1952 B. New Species and Genera of New Zealand Land Snails, with a Revision of the Genus Cavellia. Dom. Mus. Rec. Zool., Vol. 2, pp. 87–97. – 1954. The Land Mollusca of Stewart Island and Solander Island Trans. Roy. Soc. N.Z., vol. 82, pp. 137–56. Suter, H., 1913. Manual of New Zealand Mollusca, Wellington, 1120 pp. Wellman, H. W. and Willett, R. W., 1942. The Geology of the West Coast from Abut Head to Milford Sound, Part I. Trans. Roy. Soc. N.Z., Vol. 71, pp. 282–306. Willett, R. W., 1950. The New Zealand Pleistocene Snow Line, Climatic Conditions, and Suggested Biological Effects. N.Z. Journ. Sci. Tech. B, Vol. 32, pp. 18–48.