540 



NA TURE 



[April 5, 1906 



is shown that the periodicity of the oscillation is given by 

 the equation 



Ian (a.T„)=l/2/«(o,/a- «/«,•), 



where m a is the period of the oscillations produced, 

 27r [,, is the natural period ol the governor, and m is the 



ratio "I the actual damping of the governor to the smallest 

 value of the same coefficient which makes the governor 

 dead beat. 



This equation of a in terms of T,, represents a series of 

 curves, some of which give rise to negative values. The 

 positive values corresponding to m=I and a,. =6-32 are 



Fig. 



shown in Fig. 1. Of these curves, only the lowest one re- 

 presents oscillations which are not evanescent. 



For this case the conclusion is reached that for satis- 

 factory running the percentage variation of speed K must 

 not be too small, and the fly-wheel effect must be designed 

 in proportion to the time lag of the governor. 



Passing to the case where two sets are working in 

 parallel, the equation giving the frequency of the oscilla- 

 tions is similar to the above, provided the damping in the 

 generators is small. In this case, however, the curves 

 which give positive values of a are quite different, being as 

 shown in Fig. 2. The straight line portions correspond 



NO. 1901, VOL. 73] 



to a = a. a , where 2717a,, is the natural period of the 

 alternator. 



Now the rate of subsidence of any oscillation contains a 

 term (1 — a ,a„), and it is consequently clear that it would 

 be unsafe to allow any value of T,, between o and 0-5, or 

 between 1 and 1.5, as shown in the figure. If, however, 

 the damping of the governor is much greater than the 

 critical dead-beat value, satisfactory working may be 



possible. 



A point of great interest is brought out in this connec-i 

 tion, viz. the influence of the fly-wheel effect. The straight 

 portions of the curves in Fig. 2 depend upon the natural 

 period of the alternator, which in turn depends upon the 

 fly-wheel. The heavier the fly-wheel the less the value of 

 a„, and consequently the longer the straight portion of the 

 curve as shown in Fig. 3. In other words, the greater 

 the fly-wheel effect the greater will be the range of T,, for 

 which satisfactory working is difficult. From this it would 

 appear that it is quite possible to provide too heavy a fly- 

 wheel. 



The article concludes with a brief reference to the various 

 methods of compounding alternators. 



GEOLOGICAL NOTES. 



TN the Zeitschrift der Gesellschaft fur Erdkunde zu 



Berlin (1905, p. 412) Prof. Dr. A. Phillipson, of Bern, 

 outlines his recent journey of 10,000 kilometres through 

 the west of Asia Minor, including Brussa, near the Sea 

 of Marmora, and Makri, on its Mediterranean inlet in the 

 south. The preliminary results indicate the existence of 

 a " Lydian mass " of granite, gneiss, and crystalline- 

 schists, which forms on the whole a hummocky country, 

 flattening itself out where the lower course of the Marauder 

 cuts into it. The inhabitants are mostly clustered along 

 the included basins of Neogene deposits. A zone of meta- 

 morphic limestones and less altered phyllites lies outside 

 litis mass, following the strike of the bow-shaped crystal- 

 line core; and the discovery of a new species of Fusulina 

 (p. 417) places part of this outer zone as Permo-Carbon- 

 iferous. To the south and south-east, the Cainozoi. earth- 

 movements have brought up folded limestones of the 

 Cretaceous and Eocene type of Greece and Rhodes. The 

 complete results of the journey will not be worked out for 

 several years. 



Dr. G. Steinmann continues, in the Berichte der natur- 

 forschenden Gesellschaft zu Freiburg-im-Breisgau for 

 September, 1905, his " Geologische Beobachtungen in den 

 Alpen." The question of Klippen-structure and overrid- 

 ing in the classical Alps of Glarus leads on to a comparison 

 with the eastern Alps. The author gives a valuable ex- 

 position of Schardt's views, which were published in 1893, 

 and which led to the conception of the breaking up of an 

 overthrust limestone mass into blocks or " klippen," which 

 lie discordantly among later sediments. Without going so 

 far as Termier (p. 32), Dr. Steinmann sees in this striking 

 theory of overfolding the true explanation of the phenomena 

 of the Biindnerschiefer and the limestone zone, and he 

 appeals to workers in the eastern Alps to consider Schardt's 

 views at least in the light of a scientific possibility. The 

 last part of the present paper includes a bold but reason- 

 able speculation as to the connection between deep-sea 

 radiolarian deposits and diabasic igneous rocks. It is 

 suggested, for the Alpine, Scotch, and other instances, that 

 these basic igneous masses accumulated under the ocean 

 floors, just as more highly silicated rocks are believed to 

 gather under continents. Consequently, a deep-sea epoch, 

 followed by one of compression and overthrusting, would 

 lead to a squeezing out of " ophiolitic " igneous rocks some- 

 where along the zone of the radiolarian cherts. 



The activity of geological research in the African 

 colonies is evidenced by a recent part of the Transactions 

 of the Geological Society of South Africa, published in 

 Johannesburg in September, 1905. Mr. A. L. Hall de- 

 scribes (p. 47) the mode of occurrence of the tin-ore in the 

 pit turesque Bushveld area forty miles north-east of Pretoria. 

 The ore was first noticed in the local granite as recently 

 as 1004, and the field was described by II. Merensky 

 in that year. Its exploitation at once followed, and Mr. 



