424 REPORT— 1869. 



experimental axles thus constructed. This suggestion of Mr. Rankine is 

 now, so far as i-egards the curvature, always followed by men of skill. . 



As a matter of fact, all makers of railway-wheels and axles, and all me- 

 chanical engineers of experience, now know that, in order to obtain an en- 

 during axle, it should be made (as shown at Plate III. fig. 4) with bold 

 hollows at the junctions of the journals with the parts on which the wheels 

 fit, and with the least possible projection at the back of the wheel-boss and the 

 formation of a curve at the point of juncture with even this small projection. 



The ordinary railway-waggon or -carriage axle has, from its simplicity, 

 been used for an illustration ; but it need hardly be said that the cranked 

 axles (those important parts of locomotives) were found to give way under 

 similar conditions of abrupt change of form. 



The writer of this paper has the very strongest conviction of the idti- 

 mately fatal eflfects of these abrupt changes. He has known instances 

 where, after many years' work in slow-going engines, large shafts have 

 broken tlirough where changes in the area of section were abruptly made, 

 although these changes of area were but to a very slight extent. The writer 

 is in the habit of introducing into his specifications the following clause : — • 

 " Bold hoUows are to be formed in the angles of all the bearings and against 

 all collars or other projections, on all shafts or axles, and generally through- 

 out the engine ; care is to be taken that every change of dimension is to be 

 made gradually." 



Even at the present day, except in the case of those intimately ac- 

 quainted with the exigencies of railway work, the necessity of attending to 

 these rules is not always appreciated. As an instance, the writer may mention 

 that, having directed the swivel-hook of a large crane to be made with bold 

 hollows, as shown at Plate III. fig. 5, the workman finished it in a man- 

 ner which he considered " a nice square workman-like job," as shown at 

 Plate III. fig. 6. The materials were all that could be desired ; but the writer 

 was compelled to reject tliis hook, because he has not the slightest doubt that, 

 had it been kept at work, it would very shortly have broken through at the 

 line xy. 



Having said thus much in relation to that which may be called the history 

 of the subject, and having brought that history to the point where it is ad- 

 mitted, by all men of skill in railway work, that, to obtain immunity from 

 accident, it is not sufficient the weakest part of an axle should be strong 

 enougli, but care must also be taken that no neighbouring part shall be 

 abruptly stronger, the writer will endeavour to s<how, in a plain, familiar, 

 and, he might say, workshop manner, some only, it may be, among many of 

 the reasons which cause abrupt enlargements to be attended with the disas- 

 trous effects we now know to accompany them ; and in doing this he will 

 consider four different propositions. 



1. That abrupt change of form is detrimental even under a quiescent 

 load. 



2. That under impact abrupt change of form is detrimental even if the 

 object sufiering the impact be supposed to be made of imponderable 

 matter. 



3. That, further, under impact abrupt change is still more detrimental 

 when the weight and inertia of the object are considered. 



4. That abrupt change is detrimental when the object is subjected to a 

 vibratory action. 



As regards the first of these propositions, viz. that abrupt change of form is 

 detrimental even under a quiescent load, let Plate IV. fig. 7 represent a 



