June io, 1920] 



NATURE 



459 



and very different from the desert condition which 

 it possesses to-day. While nothing was found 

 in Palestine of the same type as the Samarra 

 ruins, the ancient Graeco-Roman temples of 

 Jupiter and Bacchus at Baalbek, in Central Syria, 



Fig. 3.— Temples of Jupiter and Bacchus at Baalbek in Central Syria, 

 showing the ground plan and some remains of later Saracenic building. 



furnished some interesting photographs. A ver- 

 tical view from about 3000 ft. gives a remarkably 

 good ground plan of the present state of these 

 beautiful remains (see Fig. 3). 



Meteorology. — The study of clouds by the 



photography from aeroplanes of their forms and 

 features has been recently discussed by meteor- 

 ologists, and need not be further mentioned. 



It would be outside the scope of the present 

 article to deal with the methods of obtaining, 

 using, and interpreting the aerial pictures which 

 have been referred to. It may be seldom possible 

 for a scientific expedition to employ aeroplanes, 

 owing to their expense ; but, when it can be done, 

 useful knowledge is bound to accrue. In other 

 cases, however, as in Palestine, photographs may 

 be taken for mapping or other purposes, which 

 will also yield important scientific material to 

 those who can make use of it; and possibly 

 photographs taken for the purpose of training 

 airmen may become of great value, even in this 

 country, if certain areas are included. Sometimes 

 the evidence furnished is clear and unmistakable, 

 but in other cases the photographs have to be 

 examined by a trained and experienced worker. 

 The general public has not been very fully 

 informed of the work of the R.A.F. photo- 

 graphers during the war, and to most people the 

 air photograph is a curiosity which seems to 

 have little value in times of peace. Though in 

 some countries the civil importance of aerial 

 photographic survey is realised, in England air 

 photography is in a somewhat languishing con- 

 dition. In these circumstances it is well to 

 remember that, though the aerial camera has not 

 been extensively employed apart from military 

 work, it nevertheless appears to have no incon- 

 siderable value in the domain of pure science. 



The Dynamics 

 By R. H, 



' I ^HE object of this article is to give a short 

 J- account of some features of the motion of 

 a spinning shell through air. Our knowledge of 

 this phenomenon has been somewhat increased 

 by war-time researches. To determine the 

 motion of a shell from the equations of rigid 

 dynamics, we require to know the complete force 

 system which represents the reaction of the air 

 on the moving shell ; but, just aS in the case of 

 an aeroplane, the components of this reaction 

 are utterly unknown a priori. The problem that 

 arises, therefore, is that of determining these 

 components by observation and analysis of the 

 actual initial motion of shells. Once they have 

 thus been determined, they can be applied, pro- 

 vided the essential conditions remain similar, to 

 the calculation of the complete motion of a shell 

 along its trajectory. 



In the simplest case of all this procedure is 

 classical. The air resistance to a shell, moving 

 so that the directions of its axis and the velocity 

 of its centre of gravity coincide, has long been 

 determined thus as a function of the velocity, and 

 trajectories have been computed assuming that this 

 coincidence subsists throughout the motion. 

 Under this assumption the problem is merely one 

 NO. 2641, VOL. 105] 



Of Shell Flight. 

 Fowler. 



of particle dynamics, of which the solution may 

 be regarded as completely known. The com- 

 parison of calculations and observations shows 

 good agreement in range and height when the 

 shells are suitable and the total angle turned 

 through by the tangent to the trajectory is less 

 than, say, 50°. The calculated trajectory, how- 

 ever, is a curve lying in the vertical plane con- 

 taining the original direction of projection, while 

 the observed positions of the shells do not lie in 

 this plane, but appreciably to the right of it when 



j their axial spin is right-handed. This well-known 



I departure from the original vertical plane is called 

 drift, and converts the trajectory into a twisted 

 curve. It cannot be accounted for on the original 

 assumption. 



It is with these cases, in which particle 

 dynamics fails to explain the observations — such 

 as the drift, trajectories of large total curvature, 

 and (as we shall see) initial motions — that we are 



1 mainly concerned here. For their study we must 

 abandon the assumption that the direction of 

 motion of the centre of gravity and the direction 

 of the axis of symmetry coincide, and study the 



! whole motion as a problem in rigid dynamics. 



I In order to do this we must, first of all, deter- 



