margin 1–2 dentate or ciliate towards the apex; dorsal lobe not crossing the stem, folded upwards on the ventral to which it is attached for its whole length, forming the keel; apex bifid, 2–4 dentate or ciliate, 0.8 mm tall including the teeth, ca. 0.4 mm wide, may be imbricate. Cells rounded-hexagonal at the apex and in the dorsal lobe, ca. 30μ, lower much longer, to 90μ, often in undulating rows visible with a hand lens, and with cell contents clustered at the tapering end walls. Underleaves bilobed, segments ± diverging 2—ciliate-dentate, cilia may be compound. Invol. leaves large, in several pairs, imbricate, progressively larger to 2 mm with laciniate, undulate-crispate margins, fertile innovations sometimes arising from their bases. Marsupium fleshy, 2 mm × 1.3 mm, tinted rose, hairy with soft pinkish rhizoids. Archegonia all at the base of the hollow marsupium, capsule valves seen as spirally arranged. ♂ branches with bracts much shorter than the ordinary leaves, closely imbricate, sub-erect, biciliate. Antheridia large, single.

B. glandulifera Col. no. a1428 is a specimen with ± flaccid leaves, with apex and ventral margin sparsely ciliate, cells large. It was collected on Mt. Tongariro by H. Hill.

Whether or not B. Hockeni Berggr. should be sunk in B. diplophylla has been a matter of long consideration. Admittedly there are the two extremes, one with ovate to broadly ovate leaves, with a 4–5 ciliate apex and with one or two cilia on the ventral margin; “foliis ciliato-dentatis … ventrali (lobo) late ovato, obtuso”. This is the original B. diplophylla (Tayl.) Mitt. Of B. Hockeni Berggren writes: “The typical form has distant or contiguous leaves with dorsal lobe very small, short, directed straight forwards, and either lobe oblong with apex almost truncate and bidentate. This form passes through intermediate ones into such as have the lobes 3–4 dentate, and the dorsal not far from half the size of the ventral”. The size of the dorsal lobe varies greatly and counts for little as a specific criterion, as do also the size and shape of the cells. If B. Hockeni is upheld, numerous specimens must go unidentified or referred to either species, as broad-leaved forms may be simply bidentate, and narrow-leaved forms may be multi-ciliate. Moreover stems with both broad and narrow leaves may be in the same specimen.

Common throughout on shady earth banks in grassland and bush from Three Kings to Stewart Island.

The type was from Auckland Island coll. Hooker. In a type fragment received from Mitten's Herbarium there was a stem of B. convexiuscula.

B. rosea Berggr. N.Z. Hepaticae, 1898.

Plants dioicous, variable, medium to large, beautiful, often rose-coloured, mostly in closely packed mats, terrestrial. Stems to 3 cm, straight, or flexuous, procumbent. Leaves to 2 mm, narrow-oblong, scarcely contiguous to imbricate, conduplicate, a little convex, sometimes decurved, ventral lobe shortly bifid at apex, margins fringed with glossy cilia from a base of 2–4 cells, carina very short, dorsal lobes imbricate, a little decurrent on the stem, rounded-quadrate, 5–6 ciliate-lobed, flat. Underleaves quadrate nearly 1 mm, bilobed, lobes ciliate or ciliate-lobed in the margins, upper ones, with the involucral leaves following the growth of the marsupium downwards Cells irregularly hexagonal at the apex to 30μ, becoming more elongated towards the base to 60μ, in undulate rows, minutely punctate, walls thin to firm. The undulate rows can be seen with a hand lens Involucral leaves in several rows, becoming larger towards the marsupium, longly laciniate-ciliate. Marsupium descending, rose-red, densely setose, 3 mm × 1.5 mm, upper part hidden by the involucral leaves. Androecia as in B. diplophylla.

Typical plants have long and numerous cilia along the leaf-margins, mostly based on 2–4 cells, while some have fewer and become intermediate between B. diplophylla and Berggren's type. A handsome plant from Jackson's, Westland, 10616 W. M., has branched cilia on the margins.

Berggren has the following note: “Stems much longer and stiffer and more equally foliated than in B. diplophylla. The pink hue and felty covering from the densely ciliate leaves and amphigastria mark this species.” Generally speaking this is correct, but the plants are not always pink, and are sometimes even smaller than B. diplophylla. Moreover B. diplophylla is occasionally tinged with pink. Apparently B. rosea does not come down to such low level altitudes as does B. diplophylla. The direction of the leaves is normally more oblique in B. rosea than in B. diplophylla.

North Island: Kaitaia, H. B. Matthews; Russell-Whangarei Rd., 1326, Bay of Islands, 676, Waipoua Forest, 1327, V. W. L.; on tracks, Waipoua Forest, 21679, 24680, Bot. Div. Herb., Hunua Ranges, H796, Waitakere Ranges, 24678, subalpine scrub and modified forest Te Moehau, L. B. M.; numerous specimens from earth banks, Trounson Kauri Park, Waipoua forest, Ruapehu, Mt. Ngongotaha, H565, Paeroa Range, 1280, K. W. A.; Coromandel Range, 790, A. L. H.; 1324, E. O. Campbell; Waitomo, 9562 (all leaves separate, light brown in packet), W. M.; headwaters of Rangitikei River, 1302, 1260, H. M. H.; Waikaremoana-Waikare-iti track (near to B. diplophylla), 9291, E. A. H.; Mt. Egmont, 1335, A. P. D.; 680, G. O. K. S.; 10184, 10164, E. A. H.; Wharite South Ruahines, 245, R. M. Greenwood; Kapakapanui, 900, Mangaroa, 693, A. P. D.

South Island: Takaka, Nelson, 673, Glen Hope, Nelson, 650, I. Haskell; South Westland, 4952, Lake Mapourika, 6547, Jackson's, 10616, W. M.; Fox Glacier, 697, Mrs. Knight; Head of Lake Manapouri to Wilmot Pass, H826, G. S.; descent of Mt. Hercules, H5608, A. M. & L. Jack.

Freshly gathered specimens from Egmont showed small oval oil bodies,-mostly at the ends of the cells.

Also recently collected in S.E. Queensland on bank in rain forest, 2,600ft, H44 Nat. Herb, N.S.W., comm. J. H. Willis.

The type was from Bealey River, with Isotachis Lyallii coll. S. Berggren.

Balantiopsis convexiuscula Berggren. Text-fig. II, Fig. 17.

B. convexiuscula Berggr., N.Z. Hepaticae, 1898.

B. aequiloba Berggr., N.Z. Hepaticae, 1898.

Plants small to large, variable, pale green in layered mats, rarely tinted. Stems to 8 cm, simple or sparsely branched, appearing convex with the leaf-apices incurved, a little flexuous, mostly lying in the same direction, apices nodding in one specimen. Leaves oval-rectangular, decurved, imbricate to 1.5 mm, apex bidentate sinus small, obtuse, ventral margin entire or with a few teeth; dorsal lobes rounded in outline, imbricate, direction straight forward or slightly oblique, medium to large, to 1.2 mm, according to the size of the plant, usually 6–8 lobate-dentate or lobate-ciliate, but often less, may be almost entire, carina variable, with dorsal lobes almost free in some cases, but tending to become longer in smaller plants Cells to 30μ, but mostly shorter, narrow, compact, apical ones small, quadrate crowded. Underleaves ovate-quadrate with an apical sinus and margins spinous-toothed. Invol. leaves with incurved margins and strongly and irregularly laciniate. Marsupium to 3 mm long × 1.5 mm wide, deeply rose, strongly setose.

This species is recognized by its rounded dorsal lobes, more dentate than ciliate, and the smaller, more compact cells. The dorsal lobes are usually flat and regular, but may at times have the margins somewhat decurved. Two specimens from Stewart Island, 480 and 411 W. M. show a tendency for the lobes to become entire.

Dr. Lyall's specimen referred to a Falkland Island species B. erinacea by Mitten in the Appendix to the Handbook also belongs here, but without further specimens from Falkland Islands, I would not go so far as to say that our B. convexiuscula is indeed B. erinacea and that Berggren's name is a synonym. Berggren (1898) stated that there was little difference between B. erinacea and B. diplophylla.

B. aequiloba Berggr. is this species. The type was submerged and it is more or less flaccid and a poor specimen.

North Island: Kaeo-Manganui Rd., 514, Bay of Islands, 1268, 10666, V. W. L.; Great Barrier Island, T. Kirk, 6068, P. R. B. Herb., wrongly identified by Stephani as B. diplophylla; shady stream bank, Waipoua Forest, H691, Trounson Kauri Park (approaching B. rosea), K. W. A.; N. W. Ruahines. 3518, A P. and H. M. D.; Mt Hector, 7478, Bot. Div. Herb., Akatarawa Saddle, 9207, Ruamahanga, 9252, V. D. Z.; common on banks and paths, Mt. Egmont, 10165, 10174, bank in bush, Mangahao Dams, Tararuas, 10636, E. A. H.; bog, Mt. Maungapohatu, Urewera, 4,600ft, 9674, 9692, A. P. D.

South Island: On rocks by small forest stream near Bealey Glacier, 100, Kelly Range, 9608, Avalanche Peak, 120, W. M.; bush near Greymouth, 690, M. Berry;