The New Zealand Forms of Prasophyllum R. Br. By Edwin D. Hatch, Laingholm, Auckland. [Read before the Auckland Institute, May 19, 1946; received by Editor, May 21, 1946.] Acknowledgments. The writer is greatly indebted to Mr. K. W. Allison, of Dunedin, whose copious notes and specimens proved invaluable. Introduction. Prasophyllum is one of the most difficult of orchid genera to classify. The superficial similarity of many of the species, and the minute flowers of others, combined with a tendency on the part of the floral segments to change their shape during development and while drying, render it almost impossible to work with dried specimens. Nevertheless a great many species were based upon such material, and the genus quickly reached a state of confusion. To overcome these difficulties, fresh descriptions from living material were necessary, and the Australian forms have been admirably elucidated during recent years by Rupp and Nicholls. The aim of this paper is to bring the New Zealand species into line with their work. The descriptions which follow (with the exception of Pr. rogersii) have been drawn up from living material collected by the writer and carefully checked against the original and any subsequent descriptions. The numbers (in italics) used in citing the New Zealand distribution refer to Cockayne's Botanical Districts. Development: Juvenile tubers bear leaves only. The tubers of the second, third, and fourth seasons bear small, few-flowered spikes, with usually long, over-topping leaf-laminae. The characteristic dense spikes are not produced as a rule, until after the fifth season. The tubers are very large in proportion to the rest of the plant. Fertilisation has been described by Dr. R. S. Rogers. The plants are adapted for insect-pollination, but should this method fail, are capable of self-fertilisation. Some of the smaller species are often cleistogamous, so that the anther connective must rise, whether the flower opens or not. The anther lies behind the stigma, with the rostellum immediately above them both. The two pollinia are attached to the modified portion of the rostellum by a linear caudicle, which is covered, in the bud, by the anther connective. When the flower opens the connective rises, exposing the upper side of the caudicle to the drying action of the air, which causes it to shrink and curl. The pollinia are rotated by this shrinking of the caudicle, and carried over to the front of the column, where they are deposited, more or less accurately, on to the viscid surface of the stigma. At any time during this recurving process the pollinia may be removed by insects.