Introductory remarks on the variableness of the species; key to 31 species; descriptions of same, including 2 Tasmanian species recently detected in New Zealand, L. amplectens and L. paucistipula, and 2 new species described in Latin, L. innovata Hodgson sp. nov. and L. fiordlandiae Hodgson sp nov., and one new variety, L. helmsiana var subparallela Hodgson var. nov, 2 new combinations, L. stronglyophylla (Tayl.) Hodg. comb. nov., and L. turgescens (Tayl.) Hodg. comb. nov.; 12 species of Stephani's definitely reduced to synonymy, also 7 of Herzog's, 1 of Schiffner's and 1 of Pearson's. Notes on Chiloscyphus and Heteroscyphus, with correction of errors in a former paper on Chiloscyphus, and new localities of rare species Thirty-five figures Types of new species in Herbaria of E. A. Hodgson and K. W. Allison.

It is with much trepidation that one undertakes the revision of a genus such as Lophocolea, and one encounters expected difficulties almost from the outset. In New Zealand the species are so variable and in such a fluid state, and the interrelations within certain groups are so manifest, that with the Old World conception of a species being what it is, it is no wonder that the number of described species has reached at least 50, not including those already synonymized by Mitten. With the number reduced to 31, as in this paper. it is hoped that the position is now less confused.

One notices early on a seeming tendency in the leaf-apices for the teeth to disappear. Even in L. bidentata, with 2 long, attenuated laciniae, there are instances in which one or even both of these have become obsolete. The matter of disappearing stipules is another difficulty, which appears to bring one subsection of the genus close to Jungermannia One wonders, too, why in some instances small cells are mixed with larger ones. as though the plant might be in process of changing from one to the other, and there seems to be no fixed rule for uniformity of cell-size in one particular species As in Plagiochila, there are numerous types of cells throughout the genus.

The originals of 9 of the species were from Auckland or Campbell Islands, collected by Sir Joseph Hooker, and to recognize the same species from the mainland, one has to allow for poor development and for possible geographical differences. The same may be said of species whose originals were from Tasmania or Australia.

Though widely distributed throughout the world, there are comparatively few species in the Northern Hemisphere, whereas in South America, Australia and New Zealand they are very numerous. L. Rodway describes 44 Tasmanian ones, but admits that they might not all be good species.

Lophocolea differs from the related genera Chiloscyphus and Mylia (Harpanthaceae) in its triangular-prismatic perianths, but this mark of distinction is sometimes hard to recognize.

Stephani has divided the genus into 3 sections, Integrifoliae, Bidentes and Heterophyllae, with numerous subdivisions in each section, with which I have not bothered.