lO 



NATURE 



[September i, 192 i 



A Geological Excursion Handbook for the Bristol 

 District. By Prof. Sidney H. Reynolds. 

 Second edition. Pp. 224. (Bristol : J. W. 

 Arrowsmith, Ltd. ; London : Simpkin, Mar- 

 shall, and Co., Ltd., 192 1.) 55. net. 

 The second edition of this useful handbook re- 

 produces the first in all essential features. The 

 author is, however, well known for his untiring 

 investigations into the geology of his district, and 

 the recent researches of his pupils and himself 

 have necessitated a number of minor alterations 

 in the descriptive portion of the book. 



The chief additions relate to the igneous rocks 

 associated with the Carboniferous Limestone, and 

 four out of five new text-figures illustrate the 

 outcrops and exposures of these rocks at Goblin 

 Combe and in the neighbourhood of Weston- 

 super-Mare. 



It is to be regretted that page-references are 

 lacking, not only in the list of illustrations, but 

 also in all the other references made to text- 

 figures. T. F. S. 



Letters to the Editor. 



\The Editor does not hold himself responsible for 

 opinions expressed by his correspondents. Neither 

 can he undertake to return, or to correspond with 

 the writers of, rejected manuscripts intended for 

 this or any other part of Nature. No notice is 

 taken of anonymous comtnunications.] 



Ruling Test Plates for N!icroscopic Objectives: 

 Sharpness of Artificial and Natural Points. 



Test plates for microscopic objectives should con- 

 sist of alternate opaque and transparent lines 

 approximately of the same width, and placed on a 

 plane surface (not grooves engraved on transparent 

 material). Considerable difficulty has been found in 

 producing such lines when the distance between them 

 is less than about 1/2000 in. They might be made 

 by ruling on thin opaque films, and, so far as opacity 

 is concerned, films of silver or other metals chemically 

 deposited on glass would meet the case ; but the in- 

 trinsic strength of these films is greater than their 

 adherence to the glass, and the whole of the metal 

 is torn away by the ruling point when the lines are 

 close together. In many trials I have never succeeded 

 in ruling on chemically deposited silver at even 2000 

 lines per inch. I have found, however, that films of 

 certain aniline colours dried on glass are well adapted 

 for the purpose, their opacity being so intense as to 

 show a fair depth of colour even when the thickness 

 of the film is a very small fraction of a wave-length 

 of visible light. Their adherence also to the glass is 

 greater than their intrinsic strength, and, so far as 

 my experience goes, the limit to the fineness of the 

 lines which may be ruled on them is not reached until 

 the spacing of the lines is less than the thickness of 

 the film. 



In ruling lines on such films the load on the ruling 

 point should be sufficient to remove the material of 

 the film, but not to scratch the surface on which it is 

 laid, and considering how soft the film is compared 

 to glass or quartz, it seemed worth while to see 

 whether a steel point might not be substituted for 

 diamond in the ruling process. 



In looking into this question, one of the first things 



NO. 2705, VOL. 108] 



to be noticed was the extraordinarily small load which 

 must be placed on the point. Suppose, for instance, 

 that it is required to rule at the rate of 100,000 per 

 inch, the area of the point in contact with the plate 

 must not be greater than (1/200,000)' in., and since 

 a grain is roughly about 1/ 15,700,000th of aton, a load 

 of I grain on the point will produce a pressure of 

 more than 2500 tons per square inch. Even hard steel 

 would not stand a hundredth of this pressure for long, 

 and though I am not aware that any accurate 

 measures have been made of the pressure required to 

 scratch glass, I should expect it to be less than 10 

 tons per square inch. 



For ruling lines at 100,000 per inch, therefore, the 

 load on the point should be not greater than i /200th 

 of a grain, and the holder in which the point is 

 carried by the ruling machine would have to be made 

 with a very small mass, and counterbalanced. 



To find out whether it was possible to grind steel 

 points to the requisite fineness, I began by examining 

 the points of needles in the state in which they are 

 sold. 



They varied in the degree of sharpness, but their 

 extremities were all somewhat parabolic in section, 

 with an average minimum radius of curvature of the 

 order of 1/20,000 in. (Fig. i). 



On trying to secure a finer point by grinding. It 

 was found that, using the lightest pressure which 

 could be applied by hand to a needle mounted at the 

 end of a light reed, the point continually broke away, 



^ Ml! ! Hill M ! I n nil! M I 



I 2 3 4-5 

 Thousandths of an inch 



liiiiiiiiiliiiiliiiiliiiil . 



12 3 4-5 



Thousan dths of an Inch 



Figs, i to 10. — {Traced from fhotograpJis ) i, needle : average point as 

 sold. 2, needle : ground under a load of 15 grains ; see broken point. 

 3, needle : softened to show bending under the same load. 4, needle : 

 ground under a load of 1 grain (about). 5, needle : dropped, point 

 downwards, on a glass plate; fall i in. 6, spine of cactus (Opuntia) : 

 • barbs much sharper than the apex. 7, thistle spine. 8, gorse prickle. 

 9, stinging-hair of nettle : this is a thin-walled tube ; a drop of poison 

 is issuing from the end. 10, bramble : immature prickle ; older prickles 

 are not so sharp. 



leaving a rough end somewhat less than 1/10,000 in. 

 in diameter (Fig. 2). On repeating the process with 

 a needle which had been slightly softened, the end 

 tended to become cylindrical, and the cylindrical part 

 broke off when its length was about two diameters 

 (Fig. 3). (This cylindrical end is analogous to the 

 "wire edge " left when sharpening a rather soft knife 

 or chisel.) 



