In 1790 Goethe (Eng. Trans., 1863) published his doctrine of metamorphosis. Before that only two authors had expressed any fundamental ideas with regard to the flower. Wolff in his Theoria Generationis (1759) said that the plant consisted of nothing but stem and leaf, the root being a modification of the former; parts of the flower were appendages, and the production of floral leaves was due to degeneration of the sap, the richest portion being used up at a lower level by the vigorous first-formed foliage leaves.

Linné (1760) expressed the same idea as Wolff, partly basing his evidence on teratological specimens, in which the various floral organs were sometimes found to be replaced by leaves. He further considered the flower to be a modified shoot, five years' growth being compressed into a single year. Believing that the different floral whorls were developed the one from the other, and observing that they encircled each other, he attributed the various floral envelopes to different tissues of the stem—e.g. the bark gave rise to the calyx, and the carpels arose from the pith.

The present work on Apseudes talpa forms the second and concluding part of an investigation undertaken in an attempt to discover from feeding mechanisms the steps by which the higher Peracarida may have been derived from a primitive Mysid-like ancestor. The first part of the work consisted of an account of the feeding mechanism of Diastylis bradyi (Dennell, 1934), but it was felt at that time that a discussion of its significance could usefully be postponed until the completion of the work on Apseudes. It is now possible, in view of the accounts of feeding in Hemimysis (Cannon and Manton, 1927), and in Gnathophausia and Lophogaster (Manton, 1928), to discuss with some measure of completeness the feeding mechanism of the Peracarida.

Our district (Pl. II) includes an exceptionally beautiful tract of the Highlands, stretching from the Errochty Water, half a dozen miles north-east of Kinloch Rannoch, to near Loch Tay. It reaches its greatest height in Beinn a' Chuallaich (2925 ft.), Schiehallion (3547 ft.), Carn Mairg (3419 ft.), and Carn Gorm (CG, Pl. II, 3370 ft.). These, with companion mountains that closely approach the 3000-foot contour, are remnants of the much dissected Highland plateau. There are two main eastwardly-draining valley systems: the more northerly holds Loch Rannoch and the River Tummel; the more southerly is known throughout its length as Glen Lyon. The southern limit of the district approximately follows the divide between Glen Lyon and Loch Tay, but fails to include the summit of Ben Lawers (3984 ft.) (inset, text-fig. 14). The eastern margin is really marked by the great Loch Tay fault; though, for purposes of comparison, some previously published mapping is reproduced on PL II east of this fault. The Tummel and the Lyon cross the Loch Tay fault at about 450 and 400 ft. above sea-level.

The work described in this paper was carried out in the Department of Zoology of Birkbeck College, University of London, and at the Marine Laboratories at Plymouth and Port Erin. I wish to acknowledge my indebtedness to the University of London for the use of their table at Plymouth, to Birkbeck College for a grant of £5 towards the illustrations, and, in particular, I would express my gratitude to Mr Graham, under whose direction the work was carried out, for his continued kindness and help.

The specimens of Lepidochitona cinereus which were used for the investigation were collected at Bangor, Plymouth and Port Erin, and also obtained from Cullercoats. Most of the material from Plymouth and one specimen from Cullercoats was infected with the Haplosporidian parasite Haplosporidium chitonis (Debaisieux, 1920) and was therefore unreliable for histological and physiological work. Specimens of Acanthochitona crinitus were collected at Plymouth, and showed no sign of infection with Haplosporidium chitonis; a few, however, were slightly parasitised with another Sporozoan. The Californian species Ischnochiton magdalenensis and Cryptochiton stelleri were obtained from Turtox General Biological Supply House, Chicago. Specimens of these species were fixed in formalin, which proved to be a poor fixative for histological work.

In a former memoir (1934) the author described the early development of the ferret from fertilisation to the formation of the prochordal plate. The following is a continuation of that work and describes the development of the mesoblast and notochord from their incipient phases to a stage with seven somites when the definitive notochord begins to separate from the notochordal plate. As far as can be ascertained these phases have not been previously described in the case of the ferret, although the literature on gastrulation and the formation of the notochord and mesoblast is voluminous. A résumé of the descriptions and theories of gastrulation has already been fully summarised by Rabl (1915), and later by Van Oordt (1921). Opinions concerning notochord and mesoblast formation will be referred to in the following pages, and as there is great confusion of nomenclature in it I have appended a Glossary of the terms which I have used and the meanings that I attach to them (p. 188).

In my account (Read, 1935) of the metamorphic geology of Unst, in the Shetlands, I have shown that the island is made up of a series of tectonic units separated from one another by dislocations of various kinds. The rocks of most of the units have lengthy metamorphic histories that are different in the different units.

This present communication deals with one tectonic unit, the Valla Field Block. The metamorphic history of this block can be divided into three stages, each characterized by the type of pressure-temperature factors operative during it. It is my main purpose here first to give the evidence, derived from the study of the pelitic rocks, on which this subdivision is based, and then to examine the behaviour of the various other rock-types during the three episodes. The results show that the stability of certain minerals depends upon the bulk-composition of the rocks containing them. In polymetamorphism of the Unst type, the minerals cannot be considered apart from the rocks in which they occur—a point of importance for Highland geology in general.

During five seasons the writer has been investigating the fishes of the Middle Devonian flagstones in the quarry at Achanarras in Caithness. This celebrated quarry, the home of Palœospondylus, has been long known and has been previously worked by Traquair (1890, 1894), who has described its fauna. There appeared, however, still to be room for a more detailed and intensive examination of the fauna such as could only be produced by a more extended period of work than had so far been given to it. By permission of Colonel Murray-Thriepland, the owner of the estate on which the quarry is situated, I have been able to spend the seasons of 1931–1935 in making a collection, of which a part is now to be described. To the Colonel and to his son Mr Peter Murray-Thriepland I have been indebted throughout for their unfailing interest and assistance. The main part of this account was written in 1934, but was not published owing to the discontinuance of a foreign journal for which it had been accepted. The season of 1935 produced some further material, which has now been incorporated. Meanwhile the writer has had the opportunity of discussing questions relating to the genus Dipterus with Dr Stanley Westoll, who has in preparation a work on the Devonian fishes in general, and is indebted to him for information on several points. Dr Westoll's view that the scales show an alternate deposition and resorbtion of the outer layer together with its bearing on questions of species is here fully adopted (Westoll, 1935).

The well-known deposits of early Upper Devonian age at Scaumenac Bay, P.Q., Canada, have yielded numerous specimens of the rather primitive Dipnoan which is now known as Scaumenacia curta (Whiteaves). During the summer of 1984 one of the authors (W. G.-S.) obtained at this locality three specimens of a very distinct new Dipnoan, characterised by an elongated head and other features. Three other specimens of this new form [No. P. 6785 in the British Museum (Nat. Hist.), the anal fin-skeleton of which was figured by Woodward (1893, p. 241) under the name Scaumenacia curta (Whiteaves), and two in the Royal Scottish Museum] were subsequently recognised in museum collections, but the present account is based entirely on the three specimens mentioned above. From these it has been possible to describe almost the whole skeleton, except the endocranium, and to compare it, in some detail, with those of other primitive Dipnoi; in this comparison we have used the results of work by Porster-Cooper (1937) on Dipterus and by one of the present authors (T. S. W., nearing completion) on certain primitive Dipnoi.