1848.T 



THE CIVIL ENGINEER AND ARCHITECT'S JOURNAL. 



309 



reason, be safely distinguished by culonred media. In many cases also, 

 where the purpose of guiding into a narrow channel is to be gained, the 

 leading lights which are used, should, at the same time, be so arranged as 

 to serve for a distinction from any neighbouring lights. 



14. Floating lights, which are very expensive and more or less nncer- 

 fain from their liability to drift from their mooring, as well as defective in 

 power, should never be employed to indicate a turning point in a naviga- 

 tion in any situatioD where the conjunction of lights ou the shore can be 

 applied at any reasonable expense." 



The building- of the lantern is a work of great care, and in 

 which our writer has very cleverly made an improvement : — 



"A considerable practical defect in all the lighthouse lanterns which I 

 have ever seen, with the exception of those recently constructed for the 

 Scotch lighthouses, consists in the vertical direction of the astragals, 

 which, of course, tend to intercept the whole or a great part of the light in 

 the azimuth which they subtend. The consideration of the improvement 

 which I had etfected in giving a diagonal direction to the joints of the 

 fixed refractors, first led me to adopt a diagonal arrangement of the frame- 

 work which carries the cupola of zones, and afterwards for the astragals 

 of the lantern. Not only is this direction of the astragals more advan- 

 tageous for equalising the effect of the light ; but the greater stiffness and 

 strength which this arrangement gives to the frame-work of the lantern 

 make it safe to use more slender bars, and thus also absolutely less light is 

 intercepted. The panes of glass at the same time become triangular, and 

 are necessarily stronger than rectangular panes of equal surface. This 

 form of lantern is extremely light and elegant. To avoid the uece-sity of 

 painting, which, in situations so exposed as those which lighthouses gene- 

 rally occupy, is attended with many inconveniences and no small risk, the 

 framework of the lantern is now formed of gun-metal and the dome is of 

 copper ; so that a first-order lantern of 12 feet diameter and 10 feet height 

 of glass costs, when glazed, about £1260. In order to give the light- 

 keepers free access to cleanse and wash the upper panes of the lantern (an 

 operation which in snowy weather must sometimes be frequently repeated 

 during the night), a narrow gangway, on which they may safely stand, is 

 placed on the level of the top of the lower panes, and at the top of the 

 second panes, rings are provided of which the lightkeepers may lay hold 

 for security in stormy weather. A light trap-ladder is also attached to 

 the outside of the lantern, by means of which there is an easy access to the 

 ventilator on the dome. 



Great care is bestowed on the glazing of the lantern, in order that it may 

 be quite impervious to water, even during the heaviest gales. When iron 

 is used for the frames, they are carefully and frequently painted ; but gun- 

 metal, as just noticed, is now generally used in the Scotch lighthouses. 

 There is great risk of the glass plates being broken by the shaking of the 

 lantern during high winds; and as much as possible to prevent this, 

 various precautions are adopted. The arris of each plate is always care- 

 fully rounded by grinding; and grooves about half an inch wide, capable 

 of holding a good thickness of putty, are provided in the astragals for re- 

 ceiving the glass, which is a quarter of an inch thick. Small pieces of lead 

 or wood are inserted between the frames and the plates of glass against 

 which they may press, and by which they are completely separated from 

 the more unyielding material of which the lantern-frames are composed. 

 Panes glazed in frames padded with cushions, and capable of being tem- 

 porarily fixed in a few minutes, in the room of a broken plate, are kept 

 ready for use in the store-room. Those framed plates are called storm- 

 panes, and have been found very useful on several occasions, when the 

 glass has been shattered by large sea-birds coming against it in a stormy 

 night, or by small stones violently driven against the lantern by the force 

 of the wind. 



The ventilation of the lanterns forms a most important element in the 

 preservation of a good and efficient light. An ill-ventilated lantern has its 

 sides continually covered with water of condensation, which is produced 

 by the contact of the ascending current of heated air ; and the glass thus 

 obstructs the passage of the rays, and diminishes the power of the light." 



We must now shut up Mr. Stevenson's book, though we could 

 very well take more from it, for it is full of new and useful 

 matter. We cannot, however, do so without giving our thanks to 

 the Board of Northern Lights for publishing this book, as they 

 before did the elder Stevenson's book on the BeU-rock lighthouse. 

 In our last number we called strongly on the engineers to write books 

 on their works ; but, perhaps, we should have done better if we 

 had called on the railway and other undertakings to find the 

 money for it, as the Board of Northern Lights have done. There 

 is something wrong now, and we cannot help thinking that the en- 

 gineers are those most to blame. If Mr. Robert Stephenson would 

 take under his care a hook on the London and Birmingham Rail- 

 way, we do believe that the shareholders would not grudge the 

 money, as each of them could have a book. They have not grudged 

 money for Wolverton, and we do not think they would for this, if 

 it were fairly put before them by their engineer. The share- 

 holders would never miss the money, while they would do a great 

 deal of good. If the engineers do not stir, we hope the share- 

 holders will ; and that we shall have hooks on our great railway 

 works, which may keep up their name and the honour of England. 



MfithemaNcsfnr Practical Men, being a common-place book of Part 

 and Mixed Mathematics, designed chiefly for the use of Civil Engineers, 

 Architects, and Surveyors ; by Olinthus Ghegokt, LL.D., F.R.A.S. 

 Third Edition, Revised and Enlarged. By Henky Law, civil engi- 

 neer. London: M^eale, 18+8. 8vo. pp. 510. 



This is a new edition of a very well known book. An editor who 

 undertakes the revision of the scientific labours of another writer, 

 undertakes an onerous and difficult task. If he alter and inter- 

 polate freely, he may be charged with disrespect towards his 

 author: if, on the other hand, he adhere too faithfully to the text, 

 he becomes responsible for the original errors, as well as for all 

 which he himself may happen to commit. 



Dr. Olinthus Gregory, of the Woolwich Military Academy, 

 published the first edition of his " Mechanics for Practical Men " 

 in 1825; and eight years after, a subsequent edition, in which ha 

 says, " I have corrected a few errors which had escaped my notice 

 in the former impression." The work commences with an elementary 

 treatise on arithmetic and algebra, and the remainder is devoted 

 to geometry and the mechanical sciences. This part of the work 

 is, in his own phraseology, " synoptical." "The definitions and 

 principles are exhibited in an orderly series, but investigations 

 and demonstrations are only sparingly introduced." So much the 

 greater, then, the importance of accuracy. Where results onlv are 

 given, the reader must trust entirely to the authority of the writer : 

 the process of investigation being omitted, there are no possible 

 means of ascertaining the accuracy of the conclusions. They must 

 be taken on trust. Like bank-notes, they may or may not repre- 

 sent sterling value ; but in the absence of direct information, their 

 circulation depends entirely on the credit of the issuer. 



Engineers at the present day are pretty well agreed, that the 

 accuracy of formulas is something more than a matter of mere 

 speculative interest — that, on the contrary, it has a real and 

 tangible importance, quite apart from theoretical considerations. 

 It was at one time thought that mathematical investigations of 

 questions of engineering were matters of mere curiosity — learned 

 pastimes. Now it is found, that if a bridge be constructed accord- 

 ing to insufficient formulae, it not only ought theoretically to fall 

 down, but practically will do so. And if the duty of a steam-engine 

 for a given quantity of fuel be inaccurately computed, not only aie 

 the laws of science infringed — but the pocket of the owner of the 

 engine suffers also. In this way, scientific accuracy comes to have 

 a practical importance, a real money value : and those who prided 

 themselves that they were " practical men," and thanked heaven 

 that they never troubled themselves about scientific theories, 

 — which were all very well for college-students, and people who 

 have nothing else to do, — discern faintly that their self-congratu- 

 lation is premature. 



These considerations render us very anxious that the theoretical 

 science should not suifer discredit, nor practical engineering injury 

 by misplaced confidence: and with this object in view, we proceed 

 to the more particular examination of the work under leview, — 

 premising that, as far as we have compared it with the preceding 

 edition, most of the errors appear to be Dr. Gregory's originally, 

 and Mr. Law's by imputation only ; and it is notliing but fair to 

 suppose that the latter was actuated by a feeling of deference 

 towards his author. 



We pass over the treatises on arithmetic and algebra without 

 examination ; being altogether elementary, they may be presumed 

 to be correct. Our criticism commences with the definitions of 

 Curves. 



" A cycloid or trochoid is ati elegant mechanical curve, first noticed by 

 Descartes, and an account of which was published by Mersenne in 1615. 

 It is in fact the curve described by a nail in the rim of a carriage-wheel while 

 it makes one revolution on a flat horizontal plane." 



Cycloids "o>'" trochoids are used as synonymous words! They 

 are names of curves essentially different : for the former, the 

 tracing point is on the circumference of the generating circle — for 

 the latter, the tracing point is within or beyond that circumference. 



" If the generating circle, instead of rolling along a straight line, is made 

 to roll upon the circumference of another circle, the curve described by any 

 point in its circumference is called an epicycloid." 



It is Jio< called an epicycloid except when the generating circle 

 is equal to the fixed circle, and rolls on the exterior of it. In the 

 other cases, the curve generated is either a Hypotrochoid, 

 Epitrochoid, or Hypocycloid. In the figure illustrating the defini- 

 tion of an epicycloid, this mistake is aggravated, by representing 

 the rolling curve as much larger than the fixed curve. 



It is of the very essence of mathematical definitions that they 

 should be precise and comprehensive ; and in no part of mathe- 

 matics is this exactness of definition more necessary than in ms~ 



24 



