In a paper entitled 'Hornfelses from Kenidjack, Cornwall' 1 the nature and origin of the rock types characterized by anthophyllite and cummingtonite occurring in the aureole of the Land's End granite were discussed.

Since these rocks were 'investigated in 1928 the writer has extended his study of them both in the field and in the laboratory, with the result that further light on the origin of these rock types has been obtained. It is the purpose of these notes to examine in more detail the evidence which is considered to add to their understanding. In the Kenidiack paper the petrography of the rocks was described at some length and in the present paper petrographic data will be given only in so far as they supplement the previous account, which may be consulted for a detailed statement.

It is now more than sixty years since Tschermak first published his classical paper on the relationship between the pyroxenes and amphiboles, but, in spite of our greatly increased knowledge of these important rock-forming minerals, many of the problems concerning their constitution and genesis remain unsolved. In particular, the nature of the conditions which determine whether a pyroxene or an amphibole will separate from a crystallizing magma are imperfectly understood, although the point is one of very considerable petrological significance. Much fuller information is necessary, particularly with regard to the relationship between the composition of the ferromagnesian minerals and the rocks in which they occur. The present paper merely represents a preliminary attempt to correlate the available chemical evidence and to determine what other possible factors might influence the character of the ferromagnesian crystallization.

The rare barium zeolite edingtonite has been examined in the course of the systematic investigation of the zeolites now being made at the British Museum, and several substitution derivatives have been prepared. The structure of natural edingtonite has also been completely determined by X-ray methods, and in the present paper an account is given of an attempt to discover the exact positions occupied in the substituted edingtonites by the potassium and thallium ions which replace the barium ions of ordinary edingtonite.

Base-exchange products obtained from some other zeolites have already been examined in detail by X-ray methods. In silveranalcime and silver-natrolite the intensities of X-ray reflections are explained satisfactorily on the assumption that each silver ion occupies the exact position previously occupied by the sodium ion which it replaces.

The first iron-rich hypersthene reported was from Vittinki in Finland and was described by M. Saxén in 1925. In 1932 N. Sundius showed that the so-called 'iron-anthophyllite' from Tunaberg in Sweden was really a very iron-rich hypersthene. He suggested that two other 'iron-anthophyllites' occurring similarly in the eulysites of Sweden at Bygdsiljum and at Mansjö might also prove to be really iron-rich hypersthenes.

Here chemical and optical data are given for the Mansjö mineral which fits well into the series and proves to be slightly richer in iron than the Tunaberg one. Data are also given for hypersthenes from Loch Duich and from Glen Buchat in Aberdeenshire, while, thanks to the kindness of Professor P. Eskola, it has been possible to obtain a sample of the Vittinki mineral and to complete the data of M. Saxén.