The following study of the oak galls mentioned by Theophrastus was undertaken at Sir D'Arcy Thompson's suggestion some four years ago. The old Greek gives no full description of any of these galls, but he characterises them so clearly in a few words that it is possible to identify all, or well-nigh all, with certainty. Böhner tried to identify them in his History of Cecidiology, but he used a poor translation and overlooked now and then the small but essential details given by Theophrastus. A fresh attempt may lead to better results.

This memoir is the result of over thirty-nine years' work among the crinoids of the Scottish Carboniferous limestones. When I began the systematic collection of these fossils few records were noted in any publications dealing with Scottish geology, and in our national museums such as those of Edinburgh and Glasgow there were, and still are, comparatively few species represented. In the official memoirs of the Scottish Geological Survey, dealing with typical Carboniferous counties like Fife and the Lothians, little more than a dozen species are recorded. Some, if not most, of these records are of doubtful value.

The highly specialised features of the Compositæ leave little room for doubt that the family is an advanced one, and it is not without significance that the vast majority of its species should be herbaceous plants; shrubby species are infrequent, while trees are of the rarest occurrence. The markedly herbaceous habit is further emphasised if attention is confined to the large and widespread genus Senecio, comprising over 2000 species, for, although they include plants of very diverse habit, few attain to tree dimensions, or possess a well-developed, perennial, woody trunk. Herbs or shrubs account for 99 per cent, of the species, the former greatly predominating. The remainder comprises a small group of about 20 species, which have assumed the arborescent habit.

The phenomena of the cyclic changes in the human and other mammalian uteri have at all times been a subject arousing interest, inviting speculation, and stimulating investigation. It is therefore somewhat curious that from the plethora of literature concerned in great measure with the histological and physiological interpretations of these phenomena relatively little information is to be obtained concerning the part played by rhythmic variations of the vascular architecture.

Forty-six years ago James Geikie published the third edition of his book, The Great Ice Age, where he expressed the view that there were five main cold periods, separated by four warmer, within the Pleistocene. Despite the enormous research by glaciologists undertaken since Geikie's time no one has yet disproved this fundamental concept; indeed, research, though perhaps casting out some of the evidence used by Geikie, has only supported his hypothesis. Yet the works of this great geologist are neglected and apparently scorned, for the terminology which he set up in plain English to indicate his five glacial periods, determined by a very wide survey, has been discarded in favour of a Continental nomenclature having only a local significance. In this work I follow Geikie in naming glacial periods, first, second, third, etc., for this has a world-wide application. It is from the writings of this master, and also those of Professor J. E. Marr of Cambridge, that I gathered the general ideas expressed here in Part I, (a), (b), and (c).

In the seaward direction in an estuary there are usually the following main gradients:—

(a) Increase in osmotic pressure (total salt content, especially chlorides).

(b) Increase in calcium and magnesium content.

(c) Increase in alkalinity.

(d) Increase in winter and decrease in summer temperature.

(e) Increase in light penetration.

There are other gradients of less importance. The first three are dependent on salinity. Thus, from the ecological standpoint, the three important physical observations relative to the study of estuarine life are the salinity, temperature, and light (silt content) gradients.

Ecological studies are directed towards the elucidation of two types of problems:

(i) Characteristics of the physical environment relevant to the distribution of specific forms of life.

(ii) The growth of mixed populations and their interrelations in a homogeneous environment or in a physical gradient.

The present study concerns itself especially with the former, and to a less extent with the latter.

In a recent paper (Gordon, 1938) reasons were given for the belief that semi-arid conditions prevailed during Lower Carboniferous times in the neighbourhood of North Berwick, East Lothian. The evidence was opposed, in a measure, to that advanced by Mr George Barrow in the East Lothian Memoir (1910) to substantiate the same position. He had relied on the absence of fossils as part proof; but, in point of fact, fossil plants have been obtained in abundance from the actual bedded ashes of Oxroad Bay that he considered (a) to be unfossiliferous, and (b) to have been formed in a manner similar to beds on the Springbok Flats of the Transvaal. The plants that have now been obtained showed xerophytic features, and, consequently, a semi-arid climate was proved on positive evidence. Other positive evidence of a lithological character was also presented in confirmation (Gordon, 1938, pp. 352, 353).

In a recent publication on the cyclic variations of the vascular architecture of the uterus of the guinea-pig (1940) we recorded that antimesometric hypersemia is a regular and essential characteristic of the uterine changes which occur during heat, and commented on its significance as a factor influencing antimesometric implantation in this animal.

A natural sequel to this work was to acquire additional evidence for our conclusions by the experimental production of antimesometric hyperæmia in the uterus of the spayed guinea-pig following treatment by injection of ovarian hormones; such evidence to be of value demands that the antimesometric hyperæmia so produced should be comparable to that of normal heat.

Between 1897 and 1903 Ardgour was investigated and surveyed for the Geological Survey by J. S. Grant Wilson. His results were published in the Summaries of Progress of the Geological Survey for 1897, 1898, 1900, and 1903, and were later included in Sheet 53 of the Survey's 1 inch to 1 mile maps and in the Glencoe Memoir. Little additional work had been attempted until 1933 when R. M. Shackelton commenced remapping the area. Shackelton was prevented from completing this work, and the writer supplanted him to continue research on the area from 1934 to 1937 under the direction of Professor C. E. Tilley.

The Ardgour estate comprises a rugged, mountainous tract of Argyllshire north of Loch Linnhe. One large and several smaller masses of epidiorite discovered by Grant Wilson (text-fig. 1) have been intruded into rocks of the Moine Series. Apart from large areas occupied by felspathic granulite and other quartzose types there are also limestones in an amount unusual in the normal Moine Series, some finely banded hornfelses and a zone surrounding the epidiorite masses composed apparently of a felspathised or granitised granulite (indicated by a broken line in text-fig. 1). Pelitic types are important in the western part of the area, and, round Sgurr Dhomhnuill in particular, they have apparently been injected by acid material. This at least is the present view, although the ground has yet to be thoroughly studied. There are a number of lamprophyre and basalt dykes, and some granite, in most places brecciated (with the occasional production of mylonite, e.g. Ben Ba), occurs along the line of the Great Glen Fault. Most of the ground is well exposed, although there may be glacial deposits inland and marine deposits round the coast (not differentiated in text-fig. 1).

The author (1928) of the present communication has already published the results of a field survey of the Lower Old Red Sandstone in Perthshire and Angus, which were linked with those derived by Campbell (1913) from his study of similar and older series of beds in Kincardineshire, to the north-east. For thirty-five miles to the south-west of the R. Tay, no detailed investigation of the sequence or tectonics of the rocks immediately adjacent to the Highland Boundary Fault, separating the Grampians proper from the Central Valley of Scotland, has been undertaken. A comprehensive account of the rocks of this belt in the region around Aberfoyle is available in the paper by Campbell and Jehu (1917).

Chironomid larvæ for a long time have been known to inhabit the brackish pools around the coast, but there are few published records of their life-histories and habits, and in many cases they cannot be correlated with their imagines. This applies equally to the fresh-water types.

The greater part of the systematic work on the Chironomidæ deals with the imagines, and much confusion exists regarding the nomenclature of species, which most authorities admit cannot be rectified until the early stages have been recorded.

The first description of a marine Chironomid larva was given by Johnston (1830), who described it erroneously as a worm under the name of Campontia erusciformis. Miall and Hammond (1900) give a good account of the life-history and habits of Chironomus dorsalis Mg., and several authors, notably Lenz (1921), Goetghebuer (1927–8–32), Thienemann (1938), and Malloch (1915), have reared some imagines from the larvae, but a very great amount of work has still to be done. Lenz (1921) has attempted to bring some order into the early stages and gives a systematic key to the genera for larvæ and pupæ, but no systematic key exists at present for the distinguishing of species.

This study was begun in 1936 primarily as an attempt to work out the life-history of the Salmon gill-maggot Salmincola salmonea (L.), a common and widespread copepod parasite of the Atlantic Salmon (Salmo salar). At that time the adult female stage only was known, although this had been reported in Scotland as early as 1766 on a kelt salmon and had been included by Linnæus among his Vermes. It was not, however, until 1939, after three years of field work, that a complete series of the stages in the life-history of the parasite had been collected and the story began to take shape. During this time not only were parasites collected, but a quantity of statistical and observational data was accumulated bearing both on the parasite.and its host. Much of this referred to conditions in the estuary, a part of the common background of the two animals which was soon recognised to be a critical region for both. Consequently special attention was given to parasitised fish from this region, and this led to a widening of the scope of the investigation to include a consideration of the habitats and interrelations of gill-maggot and fish, indicated by the term “ecology” in the title.

Since 1912 I have had the intention of re-examining all the available material of palæozoic amphibia, and have done so personally in the case of the Labyrinthodontia, and through my students Drs Bulman, Whittard, and Steen for the Branchiosaurs, and all the small animals from the famous localities of Czechoslovakia; Linton, Ohio; and South Joggins, Nova Scotia. But these researches, although they have included as complete an account as can be made of Hylonomus lyelli, the first animal called a Microsaurian, still left us without a clear conception of that group, if indeed it has any real existence.

Last year, however, Dr Steen was able to show that Microbrachis and Hyloplesion from the Upper Coal Measures of Nýřany were related to Hylonomus and were Adelospondyli.

It therefore became important to examine the well-preserved skeleton, from the Coal Measures of Mazon Creek, Ill., which R. L. Moodie described and restored as a Microsaur. The unexpected results form the paper which follows.

During the last few years the distribution of the macrofauna, and to a less extent the microfauna, of the intertidal soils of the shores of Britain has received considerable attention and the principle that the species show a zonation with maximum densities at certain levels is now well established. The extensive studies of Elmhirst (1931) and of Stephen (1928, 1929, and 1930) on the fauna of the sandy and muddy areas of numerous Scottish bays has laid the foundation for much further work in this area. Pirrie, Bruce, and Moore (1932) have described the zonation of the fauna in the intertidal sand of Port Erin bay. Alexander, Southgate, and Bassindale (1935) refer to the fauna of the sand at the mouth of the River Tees, but mainly from the point of view of comparison with the fauna of the estuary. Crawford (1937 a and b) gives notes on the distribution and habits of certain intertidal species in sandy bays and estuaries in west England and south Wales. Rees (1939) has surveyed in outline, as part of a study of the interstitial sand copepods, the macrofauna of some sandy bays in north Donegal. Spooner and Moore (1940) have given a detailed description of the zonation of the species in the estuarine muds of the Tamar. Apart from the above, reference must be made to two intensive studies on the soil fauna of the eastern shores of the North Sea, namely, that of Thamdrup (1935) on Danish shores and of Wohlenberg (1937) on certain bays on the island of Sylt. A general discussion of the ecological problems of sand, mud, and algal environments is given by Remane (1933), based on the macrofauna and microfauna of Kiel bay. He reaches the conclusion, based on the number of species which occur in each environment, that the macrofauna of sand is comparatively poor compared with mud.

The Bin and Clashindarroch Forests lie some 40 miles to the north-west of Aberdeen. Clashindarroch Forest, the larger of the two, stretches from the Rhynie-Cabrach road in the south almost to the Huntly–Glass road in the north. The Bogie valley in the east forms another rough boundary for the hill mass that the forest occupies. On the west the area marches with the county boundary. The Bin Forest consists of two main sections—the Bin itself to the north-west of Huntly, and the Balloch which lies between Keith and Grange, being partly in Banffshire. Both of these areas are more or less isolated hills.

This section under the above designation was first established by Pax in Engler's Bot. Jahrb., x, 205 (1889). The species chosen as representative of the section was P. nivalis Pall. This species with its numerous and widespread immediate allies affords an admirable centre for an exposition of the members of the section. The original definition of the section is still reasonably adequate, but its clarity was somewhat obscured by the admission of several species which indubitably belong elsewhere. Of the nine components quoted by Pax, five must be removed—P. sikkimensis and P. secundiflora, as well as the American P. Rusbyi and its two associates. In his Monograph (1) published in 1905 Pax made certain additions and corrections. The number of species indicated as within the section was increased to 15 He removed P. Rusbyi to what is now Candelabra, but P. Cusickiana and P. angustifolia were retained; so also were P. sikkimensis and P. secundiflora. On good grounds were included P. pumila, P. Aitchisonii, and P. eximia, but dubiously P. pulchella and P. Prattii. Correctly in our opinion came in P. Maximowiczii and its two allies P. szechuanica and P. tangutica.

Elsewhere two of the present authors (Davies and Francis, 1941) have expressed the opinion that a specialised cardiac conducting system, consisting of nodal and Purkinje fibres, is a newly evolved one in birds and mammals. The general morphology and topography of this system in the heart of the bird (Davies, 1930 a, 1930 b) is similar to that in the heart of the eutherian mammal, and we have postulated (Davies and Francis, 1941) that the system has undergone parallel evolution in these two classes of homoiothermal vertebrates in response to functional requirements, and that in particular its presence can be correlated with the rapidity of the heart-rate in proportion to its size. On the other hand, Keith and Flack (1907), Keith and Mackenzie (1910), and Mackenzie (1913) maintain that the sinu-atrial node, atrio-ventricular node, and atrio-ventricular bundle of the mammalian heart are remnants of more extensive tissues of similar structure in the hearts of lower vertebrates. They trace the evolution of the conducting system of the mammalian heart from a simpler and more definite form which they described in the fish, and state that as one ascends the animal scale the concentration and reduction of nodal tissue becomes more marked. We ourselves (Davies and Francis, 1941) have failed to find any nodal or Purkinje fibres in any part of the hearts of the following cold-blooded vertebrates: fish (eel, gurnard), amphibia (salamander, frog), and reptiles (tortoise, Mississippi alligator, Sphenodon). We therefore believe that, whereas the ordinary cardiac muscle is able to conduct the impulse for cardiac contraction quickly enough in these slowly contracting cold-blooded vertebrate hearts, a specially developed system is evolved in the much more rapidly beating hearts of birds and mammals. Amongst mammals themselves, the nodal and Purkinje fibres of the monotreme heart (Davies, 1931) have the same general disposition as those in the eutherian mammal. The present work is concerned with the conducting system of the heart of the remaining sub-class of mammalia, namely the Didelphia. No previous detailed study of this system in the marsupial heart has been made.