TKANSACTIONS OK SECTION A. 551 



those who see two, diuliromic ; and the totally colour-blind, monochromic. 

 There are many defjrees included in the dichromic class. There may or may not 

 be a neutral band, and this is widest in those cases approaching most nearly to 

 total coloiir-blindiiesi. 1 have recorded a case of a patient who was colour-blind 

 with one eye. It is an interesting^ fact that for form vision the colour-blind eye was 

 much the better of tlie two, and he could recognise line lines in the spectrum with 

 this eye wliich were not \isible to the other. He saw the two ends of the spec- 

 trum tinged with colour and the remainder grey. It will be noticed that his colour 

 sensations were limited to the extreme red and the extreme violet — namely, those 

 colours which present the greatest physical contrast to each other. Neither the 

 red nor th" violet appeared of the nature of a primary colour, but gave the 

 impression that they were largely diluted with grey. A theory of colour vision 

 must account for a case of thi^ kind and also for the other varieties and degrees 

 of coluur-blindnesd. The trichroiuic are a very important class, and any theory 

 must account for the fact that they see yellow as red-green and blue as violet- 

 green. As we should theoretically expect, when there is shortening of the 

 spectrum the centres of the colours are moved towards the unshortened side. 



I will conclude by showing how this theory will explain the trichromatism of 

 normal colour vision. It also explains why certain persons see spectral yellow as 

 red-green and spectral blue as green-violet. In past ages all saw the rainbow made 

 up of only three colours — red, green, and violet. When a new colour appeared 

 between the red and green (yellow) it is obvious that a mixture of red and green 

 would give rise, not to red-green, but to the colour which had replaced it — namely, 

 yellow. 



11. Lifjld Aluminium Tubes. By F. H. Nalder. 



When engaged in experimental work many are confronted with the difficulty 

 of obtaining stitt' pointers, which at the same time are light. In the production of 

 commercial measuring instruments this same difficulty has arisen, and makers are 

 being compelled to lind other means for making pointers than the old one of 

 putting a flute on a piece of thin flat metal. The method generally employed is to 

 bend np a piece of thin metal in the form of a tube with a loose .<<eam ; bnt a 

 better plan is to use seamless drawn aluminium tubes which can be made with 

 walls having a small thickness, such as one-thousandth of an inch. The specimens 

 submitted vary from this thickness with different diameters np to three-thousandths 

 of an inch or more. 



Department II. — Astronomy and Cosmical Physics. 

 The following Papers were read : — 



1. Exhibition of Celestial Photograp]i,s from the Yerkes Observatory, 

 By A. R. HiNKS, M.A. 



2. Possible Chanyes on the Luiiar Surface. By 8. A. SauNder, 21. A. 



In 1864 a strong committee of the British Association was appointed for the 

 purpose of ' Mapping the Surface of the Moon.' In their ffrst report they laid 

 stress on the necessity for thoroughly remapping and describing the surface, 

 inasmuch as ' in the present state of selenography we cannot say of any object, 

 " It is new." ' Tiie whole or parts of the surface had been previously mapped or 

 drawn by Ilevelius, Eiccioli, Dominic Cassini, Schroter, Lohrmann, Beer and 

 Miidler ; but whilst there were considerable differences between them, even in the 

 larger formations, the minute detail was very inadequately represented by all. In 

 1866 Schmidt announced that Linn6 had lost the crater form with which he had 

 previously been acquainted, and was to be seen only as a small white cloudy spot. 



