THE CONSTITUTION OF THE COPPER-TIN SERIES OF ALLOYS. 17 
Most of what we have said above is a matter of very general knowledge to students 
of metals, but as the argument of the present paper depends largely on the interpre¬ 
tation of the patterns seen in the etched surfaces, we have thought it well to state 
the more obvious rules for interpreting such patterns. The uniformly oriented 
scaffolding or grain is by some writers called a “ crystal.” We, on the whole, prefer 
the term “ grain,” or the more descriptive term, “ crystal skeleton.” - Crystals of 
considerable size, with plane faces, are sometimes found in alloys; one at least of the 
substances described in the present paper is found in such ciystals , we have also seen 
them in alloys of aluminium with nickel or with platinum.''^ 
The Development o f Pattern. 
When there is a difference between the chemical composition of the combs and the 
matter surrounding them the pattern can he developed m several ways. If there is a 
difference in hardness, one of the two materials will be etched out by polishing and a 
pattern formed in relief. The pattern thus obtained is often very well seen when the 
surface is examined with a lens, but in the CuSn alloys we have not found it suitable 
for microscopic examination or photography. A better method with all the bronzes 
ranging from pure copper to the alloy with 25 atomic per cents, of tin, is to heat the 
polished surface cautiously by laying it on an iron plate over a small flame. The 
copper-rich parts of the alloy then oxidise more rapidly than the tin-rich parts, the 
tints orange, red, blue, white, red and liliie appearing in succession m such a way that 
the copper-rich parts are always a tint ahead of the tin-rich. Very splendid patterns 
are thus obtainable in the case of the ABC alloys, and the method is most valuable in 
determining the parts richest in copper. Behrens and Stead have Iwth drawn 
attention to the applicability of this method of heat oxidation in the study of the 
bronzes. The chilled alloys, however, are in an unstable condition, anci we have 
sometimes found that the heat needed to bring out pattern by oxidation pioduced 
changes in the structure of a chilled alloy. It is, therefore, best to commence the 
examination of a chilled alloy by a chemical method of etching. 
Many re-agents are available for etching, the most suitable for a particulai alloy 
depending on the position of that alloy in the series. Ammonia readily attacks, 
dissolves and darkens the copper-rich a combs of the ABC alloys, leaving the 
remainder of the surface, as a rule, a dead ivory white, and wheie, as in unchilled 
alloys of the region CD, there is a, or a copper-rich phase resembling a, that also is 
attacked. A prolonged exposure to strong ammonia generally brings out a striation 
in the yS; ammonia does not touch alloys containing more than 20 atomic per cents, of 
tin. Strong hydrochloric acid is similar in its action to ammonia on tlie alloys fiom 
* What has been said above as to the formation of crystal skeletons does not preclude the possibility of 
the phenomenon, observed by Lehmann, Quinke, and others, that the solid skeletons may be preceded 
by the separation of droplets of an unstable lirpiid phase; the highest chills of the DE alloys stionglj 
suggest this (fig. 54). See also the note, p. 19. 
VOL. CCII.—A. T> 
