August 14, i%M~\ 



NA TURE 



361 



His first principle, that the primary object of every county 

 museum should be to make as perfect a collection as possible of j 

 the natural productions of [that county, exhibited in the most 

 attractive form, is undoubtedly correct, and has been definitely 

 sanctioned by a large body of competent authorities. This is 

 just what can be done by each provincial museum better than by 

 any other institution in the world, and it is just what is especially 

 wanted by scientists, by students, and by the public. But this 

 unfortunately has not been made the primary object either at 

 Leicester, Nottingham, or Derby. 



Mr. Sharpe's second principle, that each museum should pos- 

 sess also a typical series of foreign specimens for comparison 

 with the local ones is equally correct. Science depends upon 

 the appreciation of similarities and differences. The faculty of 

 careful observation and comparison is of such fundamental value 

 in all education and might be so effectively cultivated in museums 

 that these institutions will in the future be found essential 

 adjuncts to the schools. But here all existing museums are par- 

 ticularly weak. Opportunities for comparison are not sought 

 out. The collections are not arranged with any special view 

 to making comparisons easy and obvious. If, for instance, the 

 local birds are in one room, and the foreign types in another, 

 comparison is made as difficult as possible. Even if they are on 

 opposite sides of one gallery, and the divisions of the orders 

 made exactly to correspond, it is still difficult. They must be 

 brought so near together that the eye can match them as it 

 would match two patterns. This can be accomplished, and in 

 an artistic and very interesting manner. It is to be hoped that 

 the museum of the future will work out Mr. Bowdler Sharpe's 

 principles much more completely than has yet been done. 



Birstal Hill, Leicester F. T. MOTT 



Measuring Heat 



During the past twelve months I have been endeavouring to 

 construct some form of calorimeter which should aid me in the 

 identification of minerals, and which should, to that end, com- 

 bine accuracy with ease of manipulation. The avoidance of 

 thermometry seemed highly desirable, and a differential method 

 suggested itself. 



One gramme of the mineral undergoing investigation, and 

 one gramme of pure silver, are heated in the same steam chamber 

 and simultaneously transferred into the muffles of two Favre and 

 Silbermann's calorimeters, made of uniform size and placed side 

 by side, similarly shielded from external sources of error during 

 experiment. The specific heats are compared by the index 

 movements of the calorimeters, the index tubes being either 

 simply calibrated or empirically graduated. 



Both calorimeters are here exposed to the same external 

 sources of error. The amount of heat lost by radiation, how- 

 ever, will, for each calorimeter, depend on the duration of the 

 experiment and the rise of temperature experienced. Now the 

 heat received is communicated to the walls of the calorimeters 

 by convection currents, ascending from the muffles, in the first 

 instance, very much more slowly by conduction. It appeared, 

 then, that an internal non-conducting shield of a porous nature 

 lining the walls and retarding convection currents might reduce 

 such loss very considerably. Experiment confirmed this suppo- 

 sition. An apparatus I have had constructed on this principle 

 by Yeates of Dublin is now nearly ready for experiment. 



J. JOLY 



Engineering School, Trinity College, Dublin, 

 August 5 



Circular Rainbow seen from Hill-top 



This is not such an unusual phenomenon. It depends of 

 course on the position of the observer as regards the sun, and 

 his "coin of vantage," viz. having a space below him. I have 

 seen it several times in my life, and remember a beautiful illus- 

 tration of it given by Mr. Bains, the artist who accompanied 

 the traveller Chapman to the Victoria Falls on the Zambese. His 

 painting was, and probably is, in the Library at Cape Town. 

 He is represented as standing on a projecting rock overlooking 

 the Falls, or perhaps I should say looking up the crevice into 

 which the water falls, and in the centre of a glorious double 

 circular rainbow. I have heard the picture much criticise. 1 and its 

 accuracy doubted, but having had actual experience of such a 

 sight, I always maintained its correctness. 



I saw lately another " bow," which struck me as very remark- 



able — perhaps because I never saw one like it before. My house 

 stands on a hill-top ; below me at some distance is a piece of low 

 ground, covered by the tide at high water. The sun was low 

 behind me, and the "bow" was formed on a mist coming up 

 from the sea and swamp. It was, however, so flat that it at 

 once arrested my attention, and I called the members of my 

 family to see a "flat rainbow"! All agreed they had never 

 seen one like it. It was quite near us, as was proved by its 

 intervening between certain objects ; but I subsequently de- 

 tected one part a long way to the left of me, showing it was, of 

 course, a true "bow," but of an enormous size. 



British Consulate, Noumea, June 17 E. L. La YARD 



THE MIGRATION OF SALMON 



DURING the last ten years some exceedingly inter- 

 esting researches have been effected by German, 

 Finnish, Swedish, and Norwegian ichthyologists as to 

 the migration of salmon on their respective coasts. Thus, 

 by careful researches, some Swedish and Finnish savants 

 have proved that the salmon, which in the summer are 

 caught in the rivers of the upper gulf of the Baltic, have 

 at another season, most probably in the winter, paid a 

 visit to the shores and rivers of Northern Germany. 

 This has been conclusively proved by salmon caught in 

 the Swedish and Finnish rivers having German-made 

 hooks in their gills and stomach. From this it is there- 

 fore apparent that, in the Baltic, salmon are in the habit 

 of quitting the rivers of Northern Sweden and Finland 

 in the autumn in order to visit the shores of Northern 

 Germany during the winter, and return to their haunts 

 in the spring. That the fish should be capable of per- 

 forming the enormous journey across the Baltic — from 

 the upper gulf to the Pomeranian coast — and back every 

 year may indeed seem incredible, but that it is impossible 

 is fully disproved by the experiments with salmon and 

 trout effected by the late Mr. Fra;ik Buckland on the 

 coasts of Scotland and England in the same direction. 



In March 1872 Profs. Virchow and Hansen were com- 

 missioned by the German Fishery Association to "mark" 

 some of the salmon which had been hatched artificially 

 near Hameln, in order to ascertain whether they were in 

 the habit of returning to the river. The fish then in the 

 hatching reservoirs were one year old, and mostly seven 

 centimetres in length, although some were twice the size. 

 Having tried cutting off various parts of the fins, it was 

 found that it was most suitable for the object in view 

 and the health of the fish to cut the so-called " fat " fin 

 right away, particularly as the fish would retain this mark 

 even when full grown. 



On March 23 and 24, 1872, a thousand salmon marked 

 in this manner were let out into the Weser. The mark- 

 ing was effected by taking the fish in the left hand, and 

 then cutting the fin away with a pair of scissors, whereby 

 the fish were perfectly uninjured. The little fat fin, 

 which is mostly found on Salmonidse only, contains no 

 nerves of any importance, and has no particular function, ' 

 so that its removal does not impair the fish in the least. 



Ever since that year the fishermen between Bremen 

 and Hameln have been on the look out for the marked 

 fish, but not until a month ago a fish was caught, weighing 

 30 lbs., at Osterdeich, just above Bremen. The fat fin, 

 which, on the fish one metre long, ought to have been six 

 centimetres, was entirely absent ; and, when the well- 

 healed cut was felt, the hard membrane indicated that an 

 operation had at one time or another been performed at 

 this spot. The fish, which was marked as a grilse in 

 1S72, was then thirteen years old— an age which in every 

 respect corresponds with the age fixed by the fishermen. 

 According to general observation, it has been demon- 

 strated that the salmon in the Weser is, when one year 

 old, from five to twelve centimetres long. In the second 

 year it has been proved that the salmon go into the 

 sea, and when they re-enter the river at four years of 



