moins.” Dendy and Olliver (1901) were the first to show that the forms without a pouch, which they speak of as at the Velasia stage, were sexually immature, while the only two pouched forms which they possessed were sexually mature or nearly so. They ascertained that the teeth of the two forms correspond in number and position, though the labial teeth of the pouched form are further apart owing to the growth of the disc between them. Finally, they recognise therefore only one species in New Zealand, Geotria australis, with three stages in its life history, namely, Ammocoetes stage, Velasia or sexually immature stage, and adult (pouched) stage. Plate (1897) described as a new genus and species from Chile, Macrophthalmia chilensis, which form he later recognised as a young stage (after metamorphosis) of his Geotria chilensis. In 1902 Plate published a systematic revision of the Southern hemisphere lampreys. Of the seven genera previously recorded from the Southern hemisphere he allows only three to stand—Geotria, Mordacia and Exomegas. Under Geotria he makes three species, chilensis, stenostomus and australis, but he himself remarks on the insignificance of the specific characters separating them. From the description and figure, I fail to see how stenostomus differs from chilensis in characters sufficient to give it specific rank. I do not consider stenostomus can stand and believe it should be included under chilensis. Waite (1901–3) quoted by Woodland (1914) states it is difficult to know in what essentials stenostomus differs from chilensis. This leaves chilensis and australis. Dendy and Olliver (1901) have shown, and this paper will further show, that these two represent different stages in one life-history. There exists, therefore, in New Zealand only one species of Geotria—Geotria australis Gray. Probably this is also the case for Australia, Tasmania and South America.

As to the genus Exomegas, it appears that this will not stand examination either. In this connection I refer again to the extract from Smitt (1901) quoted above. Further Lahille (1915) says “El género Exomegas es uno de los tantos nombres que hay que suprimir de una vez de la sistemática, y hay ostros zoólogos quienes piensan lo mismo como yo. Smitt, por ejemplo hablando de los caracteres de Exomegas y Geotria, dice con razón que no corresponden, à divergencias de forma, pero si a simples grados de desarrollo, y que no merecen por cierto ser utilizados para establecer divisiones genericas” and again “Se puede decir que los ejemplares que se obtienen, se determinan como Geotria australis cuando se encuentran en los rios del centro y del sur de Chile o bien en Australia. Reciben, al contrario, el nombre de Exomegas macrostomus cuando se obtienen en la Patagonia argentina desde el río Gallegos al rio Colorado y aun hasta el rio de la Plata o la isla de Flores (Republica Oriental del Uruguay)!” Plate (1902) does not discuss Exomegas. Finally, the genus Mordacia is credited with three species. Of this genus I know nothing, since it has never been reported in New Zealand. This appears strange, since its occurrence is reported in Australia, Tasmania and Chile. Regan (1911) recognises two Southern hemisphere genera—Mordacia and Geotria. Exomegas is placed as

a synonym under Geotria. But under Geotria he actually lists five species—chilensis, stenostoma, saccifera, australis and macrostomus. But the characters used by him as specific will not bear examination in the light of a knowledge of the life history of Geotria. These characters are nearness or spacing of the labial teeth—presence or absence of gular pouch—smallness or largeness of disc—two cusps or three cusps on anterior lingual lamina—small changes in relations of fins—small changes in cusps of supraoral lamina. Dendy and Olliver (1901) had already shown that certain of these characters were liable to change at different stages in the life history. This phenomenon will be discussed and confirmed more fully in this paper.

The latest paper dealing with the Southern hemisphere lampreys is that of Lahille (1915). He concludes, “Creo, sin embargo, y sin hacerme ilusion sobre el valor de la palabra especie, que se pueden considerar como distintas especies à G. chilensis y a G. australis,” and then follow his reasons, which briefly are—(1) differences in length of the two forms; (2) differences in height of the two dorsal fins; (3) differences in the interval between the dorsal fins; (4) differences in the interval between second dorsal fin and caudal fin; (5) differences in the relation between the length of the snout and the length of the head. These differences exist, but they characterise different stages of the life history and are to be explained as a result of the shortening which almost certainly takes place with the approach of and during sexual maturity. Phillipps and Hodgkinson (1922) among the edible fishes of New Zealand list the lamprey—one species —Geotria australis. In a later paper, Phillipps (1927) lists two species of lamprey—Geotria australis—Southern lamprey, and Geotria saccifera—Northern lamprey. In a still later paper, however, Phillipps (1927A) lists only one species—Geotria australis. As I have indicated above, and as I hope this paper will show, a careful study of the life history justifies the statement that there is only one species of Geotria in New Zealand—Geotria australis.

During the course of this work, I received two important papers by Cotronei, which have considerably influenced my opinions. The first (1927) dealing with the insular organ of Petromyzon, will be discussed later when treating of a similar organ in Geotria. The second (1927A) and a previous note of his (1926) suggested the clue to facts, which have puzzled most students of the species of Geotria—namely the differences in total length of the forms studied and the differences in the intervals between the fins. In his paper Cotronei showed and was able to verify experimentally, that in the species studied by him, sexual maturity is accompanied by a process of shortening, and by changes in the intervals between the fins. His work will be referred to more fully in the text.

I take this opportunity of expressing my thanks to the following gentlemen for their invaluable assistance: Mr. W. J. Gray (Okato), Mr. B. C. Lysaght (Hawera), Mr. T. W. Downes (Wanganui), Mr. L. S. Mackie (Otakeho), Mr. Topine Ngatai (Taumaranui), Mr. R. Hatrick (Wanganui), and to Professor H. B. Kirk. I feel par-