Sept. 5, 1872] 



NATURE 



387 



by a vei7 intelligent Jew residing in Morocco, so far as they rest 

 on perS'jnal knowledge, deserve more confidence. 



The following are the chief points as to which T think myself 

 entitled to express an opinion, premising that as to some of them 

 I may place undue confidence in my own personal conclusions ; — 



1. The portion of the Atlas chain that is seen from the city of 

 Morocco is considerably higher than has generally been sujiposcd. 

 The hiijher summits approach neaily to the same elevation, and 

 the majority of these approach very nearly, if they do not occa- 

 sionally surpass, the level of 14,000 feet. Westward of the dis- 

 trict of Glaoni, S.W. of the city of Morocco, the range subsides 

 gradually as it approaches the coast. 



2. There is a certain amount of tolerably good evidence tend- 

 ing to show that the interior part of the range extending from the 

 upper valley of the Wed Tessaout to Eastern Morocco contains 

 peaks of higher elevation than any seen by us. 



3. The existence of an anti- Atlas or range parallel to the main 

 chain, and enclosing on the south side the great valley of the 

 Sous, was established by Rohlfs, if not by previous travellers ; 

 but we are probably the first who have looked across the wide 

 intervening space and scanned the outline of the anti-Atlas. 

 The portion seen by us at a distance of from 50 to 5o miles is 

 far less bold in form than the main range. The utmost height of 

 that portion can scarcely exceed 10,000 feet. 



4. The map, compiled by Capt. IJtaudouin, and published in 

 Talis at tlie Hepot General de la Guerre in 1848, which is de- 

 cidedly the best that has hitherto appeared, is defective in repre- 

 senting lire mam chrdn as arising abruptly from the low country, 

 scarcely indicating considerable lateral valleys. At the same 

 time it should be ri marked that the projecting ridges which 

 divide these la'eral valleys at pear to be lower in comparison with 

 the peaks of the main chain than is usual in other great mountain 

 ranges. 



5. There is a marked tendency to the forrrration of consider- 

 able valleys parallel to the main chain, and in such cases the 

 remark made in the last paragraph does not apply. -Some of 

 the higher peaks, and amongst them those named Miltsin by the 

 late Captain Washington, lie in ridges nearly parallel to the 

 main chain. 



6. It appears at least possible that the Anti- Atlas, if we may 

 so denrrninate the range forming the southern boundary of ihe 

 Sous Valley, is merely an example on a large scale of one of 

 the parallel ridges just referred to, many examples of which are 

 to be found in better known mountain regions. 



7. The existence of two parallel chains so continuous as those 

 represented in Gerhard Rohlfs' map appears to be open to reason- 

 able doubt. In the absence of direct evidence, it appears at least 

 equally probable that the conformation of the main chain may 

 be best represented by a series of ridges slightly inclined to the 

 axis of elevation of the entire mass. 



8. The remarkable valley of the Beni mquald, laid down on 

 lieaudouin's map as extending more than one hundred miles 

 from S.E. to N. W. in a nearly direct line must be pronounced 

 imaginary or based on false information. The details given irr 

 Rohlfs' "Reite durch Marokko," however incomplete, are nrani- 

 festly inconsistent with the general I Ian of the mountain S}sttm 

 laid dcwn in that map. 



SECTION G— Mechanical Science 

 Expcnntciils on Surfice Fikliou in Ifa/cr, by W. Froude, 

 F.R.S. 



The olj<.cti'f investigation was to determine the laws which 

 go\ern this ,oice, in those especial relations under which it forms 

 a portion of the resistance experienced by a ship when movir g 

 through the water at various speeds. 



These are, (i) the relation of the force to the speed, (2) its 

 relation to the quality of the surface, (3) its relation to the kngth 

 of the surface along the line of motion. The necessity of in- 

 vestigating it under the latter of these relatii ns, may not be at or.ce 

 obvious, it having been generally held that surface friction varies 

 directly as ihe area surface, and will be the same for a given area 

 whether it be long and narrow or short and broad. liut a little 

 reflection shews that this cannot be so, because the portion of 

 surface that goes first in the line of motion, in experiencing 

 resistance from the water, must reciprocally communicate to the 

 water motion in the hne in which itself is movirg, and, conse- 

 quently, the portion of sur/ate whiih sircceeds the first, must be 

 rubbing, not against stationary water, but against water partially 

 movirrg with it, and cannot experience as much resistance from it. 



The experiments were performed with carefully made ajipara- 

 tus, which automatically recorded the resistance and the speed, 

 and the errors and uncertainties of the results probably did not 

 in any case exceed on the whole ,| per cent. 



The surfaces used in the experiments were of yellow pine 

 board, about v'u of an inch thick, loaded at the edge with lead 

 keels of the same thickness, fastened fair and flush with the 

 board, the weights being such as to just ntutialise the flotation, 

 and hold the boards ttably in a vertical plane. The he.id end 

 (so to call it) of each board in turn was fastened into a tin sheath, 

 or fine-edged cutwater, which formed a portion of the dynamo- 

 metric apparatus, and which held the board resolutely in a ver- 

 tical plane, with its length horizorrlal, and in the line of motion. 

 The width of each board, including the lead keel, was 19 in., 

 their lengths, including the cutwater, varied from i ft. to 50 ft. 

 Great pains were taken, and successfully, to eliminate, and, in- 

 deed in eflfect to obliterate the re.'-istance due to thickness. 



It turned out that the effects of the three conditions under 

 which the variations of the force were to be determined, could 

 not be regarded as absolutely independent of each other, because 

 certain variations in the quality of the surfaces were found to 

 afl'ect in some degree the relations of the force to the spceel and 

 to the length. The results may be approximately stated in brief, 

 as follows : — 



1. As regards the relation of resistance to speed. With the 

 surface coated with shellac varnish, May's composition, or Pea- 

 cock's composition, or tallow, the resistance varied very nearly 

 as the power 1-83 of the speed ; with the surface coated with 

 tinfoil, very nearly as the power 2 05 of the speed ; but the ex- 

 perinrents with the tinfoil are not yet conrplete. 



2. As regards the relation tf resistance to quality of surface. 

 With the surface coated with shellac varnish. Hay's composition. 

 Peacock's composition, or tallow, the resistance differed ex- 

 tremely little; such variations as occrrrred scarcely exceeding 

 I per cent., and being probably not greater than belonged to the 

 small differences of smoothness in the laying on the composition. 



With the surface coated with glue, and thus simulating the 

 sliminess of a living fish, three succcs^ive experiments were tried 

 at the same speed, so as to test the effect of the gradual growth 

 of the slimy character. The first ex\ crin ent showed an inciease 

 in resistance of 2 per cent., the last of 4 per cent,, as compared 

 with the shellac surface which the g'ue resembled before immer- 

 sion, a proof that the attempted imitatien of the fish's surface 

 was not advantageous. 



Comparing a tinfoiled surface with one ccated with shellac, 

 when the length is i ft. the resistance of the former is on the 

 average only | that of the latter, making the comparison with 

 planes of 1'6 in length, the ratio is J, and with planes of i6ft., 

 more than ^\, instead of | ; indeed, the total difference becomes 

 progressively less as the planes compared are longer. At higher 

 speeds also the difference tends to become less, in consequence 

 of the higher power of the speed to which it is proportioned 

 with the tinfoiled sur'ace. 



3. As regards the relation of resistance to length of surface. 

 There plainly is a very ccnsiderable diminution of average re- 

 sistance per square foot as the lengih of surface is increased, and 

 this probably fiom the ceurse already indicated, though the raU- 

 of diminution becomes gradually lets as the surface becomes 

 longer ; there is, in fact, as gieat a diminution between 3 It. and 

 4 ft. of length as between 30 and 50. 



The following tabular statement gives the mean resistance per 

 square foot on surfaces of from i foot to 50 feet in length, with 

 speeds of from 2C0 to 800 feet per mir.ute. 



The table is applicable to a clean planed surface coated either 

 'ith shellac varnish, Hay's or Peacock's composition, or tallow. 



