632 



NATURE 



[October 22, 1908 



from below they appear black, whereas if you look at them 

 from above they still remain honey-coloured. The accom- 

 panying diagram (Fig. 2) shows to demonstration that this 

 can only be the case if the cones lie on an inclined plane 

 with the apices a little above the plane which divides the 

 cornea horizontally into two equal halves. 



Thus the simple mechanism by means of which the 

 change of colour is controlled by the spider (in at least 

 si.\ species of spiders, and most likely in many others) has 

 been satisfactorily e.xplained. 



T. Padmanabha Pillai. 



A. V. Lodge, Main Road, Trevandrum, Travancorc, India. 



Meteorology of the Indian Ocean. 



The immortal Bacon laid it down as an axiom that the 

 scientific man loveth truth more than his theory. Let 

 this serve me as an apology for trespassing on your space. 

 In Nature of September 17 there appeared an interesting 

 review of a work by the Netherlands Meteorological Insti- 

 tute, dealing with the meteorology of the Indian Ocean, 

 which appears to me deserving of comment. Your re- 

 viewer says that the data therein are made available " for 

 the sailor and the meteorologist." Permit me to suggest 

 that this statement is unfair to the sailor. Every ship- 

 master is perforce a meteorologist, but every meteorologist 

 is not a shipmaster. Much, however, depends upon what 

 is meant by the ink-horn term " meteorologist," and it 

 is necessary to define our terms before proceeding to 

 argument. 



The reviewer credits the Netherlands authorities with 

 having omitted from their charts " unnecessary details or 

 results of doubtful utility," including fog. I would 

 submit that information with respect to fog probability 

 is both necessary and useful for seafarers. Some ship- 

 masters might stoutly assert that such information is to 

 be preferred before average isobars. Rain gauges are not 

 usually carried on board ship, and to calculate humidity 

 from the large majority of the wet- and dry-bulb observa- 

 tions taken at sea would be but to court confusion. The 

 readings of rain gauges and of wet bulbs on board ships 

 under way are often destitute of scientific value by reason 

 of the environment. Hence, in my opinion, the Nether- 

 lands charts gain in quality by the decrease of quantity 

 under those heads. 



The reviewer emphasises the contention that the average 

 sea-surface current, in regions where the wind is fairly 

 constant in direction, flows to the left of the average 

 direction of the surface wind in the southern hemisphere. 

 Surely such deviation is merely apparent ! Current charts 

 of the several nations differ inter sc. Perhaps too much 

 has been made of the effect of the earth's rotation on 

 the direction of the surface wind, and your reviewer pushes 

 the theory off the solid ground of nature. With one dip 

 of ink the meteorologist dwells on cyclonic indraught and 

 anticyclonic outdraught as though the earth were at rest, 

 and with the next dip he fixes our attention solely on the 

 rotation of the earth, to the utter exclusion of indraught 

 and outdraught. One thing is certain. Had the reviewer 

 hi'Iped to make ocean currents by faulty steering, or dealt 

 with the current data in ships' log-books, it is probable 

 that he would be chary of making any hasty generalisa- 

 tion even though it appear to agree with Nansen's drift 

 or Ekman's theory. I seem to remember that serious 

 objections to the geographical positions of the former 

 remain uncontroverted. 



Many of the so-called currents used in the compilation 

 of the Netherlands charts are open to re-consideration. 

 On p. II, for example, the components of three alleged 

 currents are -I-5 and —6, with a resultant of 26 miles in 

 twenty-four hours. Either the currents were insignificant 

 or they were contradictory. Had every difference between 

 a ship's position by observation and by dead reckoning, 

 which gave a resultant of five miles or less, been regarded 

 by the compilers as " no current," as is usual in this 

 country, a large proportion of the Netherlands current 

 resultants must have suffered modification. Errors in 

 steering, variation, deviation, leeway, distance run, and 

 other items which are known to the sailor, are all dumped 

 into the dead reckoning position, and even the position 

 by observation of a heavenly body is not free from imper- 



NO. 2034, ■^'OL. 78] 



fection. Moreover, the seaman has a way of his own in 

 keeping a log-book, and unless the log-books used by the 

 Netherlands authorities were first carefully examined by 

 nautical experts, the results obtained are probably mis- 

 leading. The sea is full of secrets, and, as Longfellow 

 sang, " only those who brave its dangers, comprehend its 

 mystery." 



The reviewer says that in the pressure charts not only 

 the average isobars are drawn, but also the average 

 pressure for each two-degree area is inserted ; but I fail 

 to follow his explanation of the latter innovation. Small 

 departures from the normal barometer values in the tropics 

 may be, as he suggests, monitors of an approaching 

 cyclonic disturbance, but, depend upon it, the seaman pays 

 far more attention to the action of his barometer, as 

 regards interference with diurnal range, for example, at 

 the instant. Granting your reviewer's argument to be 

 sound, there does not appear to be any logical connection 

 between it and the erratic differences of the individual 

 means on the charts from the isobars. The Netherlands 

 observations, as is evident from previous publications, are 

 on zones which join the Cape with Sunda Strait and Perim 

 with Acheh Head. Elsewhere the results are on an un- 

 satisfactory foundation. In one two-degree area, for 

 example, there may be a discontinuous series of six observa- 

 tions, in 1856, 1886, 1887, 1897, 1903, and 1904, re- 

 spectively, nearly all of gale force ; and in the adjacent 

 rectangle six times as many observations, all in one year, 

 from a single ship detained by light winds and calms. 

 This explains the fallacy involved in the reviewer's infer- 

 ence. 



Dr. Shaw, F.R.S., in his address at the British Associa- 

 tion gathering in Dublin, pathetically referred to the ever 

 rising tide of meteorological literature, and many are 

 overwhelmed by it. Byron has well said that " a man 

 must serve his time to every trade, save censure, critics 

 all are ready made," but the above remarks are written 

 to elicit the truth, and a long apprenticeship has been 

 served by me in marine meteorology. 



Wm. Allingiiam. 



" Saratoga," Clairview Road, Streatham, October 3. 



Mu. Aluxgham submits that information with respect 

 to fog probability is both useful and necessary to the sea- 

 farer. I do not deny the usefulness ; what I implied was 

 that in the present state of meteorology no adequate in- 

 formation of fog probability could have been conveyed to 

 sailors by making charts of fog from the observations 

 for the region under consideration. The charts showed 

 the northern limit of the region in which the sailer might 

 expect fogs, and this was stated in my article. 



I was aware of the fact that rain gauges were not 

 usually carried on board ship, and that observations of 

 the wet-bulb were not always made with the carefulness 

 and precision necessary for humidity calculations. It 

 was to stimulate interest in these important meteorological 

 elements that I remarked on their absence from an 

 important publication. It is not impossible to obtain 

 really useful results for rainfall and humidity by observa- 

 tions made at sea. 



With regard to the ocean currents, I was directing 

 attention to , a point worthy of further investigation. 1 

 did not push the theory anywhere. I commented on the 

 fact, obvious on a comparison of the charts, that the 

 wind and current were related in the way indicated by 

 Ekman's theory. 



It is useless for Mr. Allingham to attempt, by directing 

 attention to the large and well-known possible errors of 

 a single observation, to abolish the cumulative evidence 

 of a long series of observations over a large part of the 

 ocean surface. I made no hasty generalisation. The 

 deviation of the current in the Indian Ocean to the left of 

 the S.E. trade wind and of the W. to W.N.W. wind of 

 higher southern latitudes is a real deviation. 



The insertion of the mean value of the barometer read- 

 ing was not intended to supplant the sailor's knowledge 

 of the effect of diurnal variation or any other effect that 

 would enable him to make a good forecast, but to sup- 

 plement it. Its value is not destroyed by the existence 

 of errors in some of the mean values. E. Gold. 



