79Q 



NA TURE 



[December 9, 1922 



which have died of old age. When the animal is 

 dead, and the soft parts rotting away, the shells, 

 being lighter than water, will thus normally rise to 

 the surface, and drift about with the surface currents, 



Fig. 3. — Chart showing occurrence of Spirula at stations of the Dana e 

 spots denote finds of live specimens, the crosses indicating stations where 

 its capture were used, but no specimens taken, 



to be eventually washed ashore. This bathypelagic 

 species, then, becomes after death a surface form, its 

 remains constituting a normal ingredient in the drift 

 of the warmer seas. 



Fig. 2 serves to show the relative size of the shell 

 as compared with that of the animal itself. The 



shell here illustrated has 35 chambers, and the 

 length (of mantle) is 39 mm. Most of the undamaged 

 shells we found on the shore were of this size. 



Distribution. — The first complete specimen of 

 Spirula known to science was taken 

 bv the Challenger near the Banda 

 Islands, west of New Guinea, and a 

 few others were captured by sub- 

 sequent expeditions (one by the 

 Blake at Grenada, W.I., another 

 by the Valdivia in the Indian 

 Ocean, and seven by the Michael 

 Sars in 1910 in the neighbourhood 

 of the Canary Islands). The chart, 

 Fig. 3, shows the localities where 

 Spirula was taken by the present 

 expedition. There were, as a matter 

 of fact, more stations than are 

 shown, but some lie so close together 

 that it was impossible to indicate 

 them on so small a chart. The 

 stations where hauls were made 

 which might have taken Spirula if 

 present, but gave negative result, 

 are marked by a cross. Our captures 

 amounted to 95 specimens in all. 



It will be seen that Spirula 

 occurred between io° and 35 N. 

 Lat. : in the eastern part of the 

 Atlantic from the Canary Islands 

 to north of the Cape Verde Islands ; 

 in the Western Atlantic from Guiana 

 and northward to the Virgin Islands 

 and Puerto Rico, throughout the 

 Caribbean ; and also in the Gulf of 

 Mexico and the Florida Straits, in the Sargasso Sea, 

 and between Bermuda and the United States of 

 America. From our previous investigations carried 

 out with the Thor, we may conclude that the species 

 is not found in the Mediterranean, or off the west 

 coast of Europe from Spain to Iceland. 



Solar Radiation at Helwan Observatory. 



'"PHE observations of solar radiation made at the 

 •*- Helwan Observatory in the years 1915 to 

 1 92 1, which have recently been published, 1 lead to 

 results of far-reaching importance. 



With regard to the standardisation of instruments 

 the position is satisfactory.. The equipment of the 

 Observatory includes two Angstrom pyrheliometers 

 made in Upsala as well as one made by the Cambridge 

 Scientific Instrument Company, which was stand- 

 ardised by Prof. H. L. Callendar. There is also an 

 Abbot silver-disc pyrheliometer. The observations 

 indicate that if a correction of plus one per cent, is 

 applied to determinations by the Upsala standard 

 it comes into agreement with the Callendar and 

 Abbot standards. A progressive deterioration in the 

 Angstrom instrument in daily use has been found, 

 which is attributed to the deposit of dust on the 

 blackened strips and a consequent reduction in 

 absorbing power. 



The usual practice at Helwan has been to take 

 several observations in the course of a morning, with 

 the sun at different heights, with the object of deter- 

 mining the " solar constant," the strength of the 

 solar heat stream outside the earth's atmosphere. 

 The usual assumption in reducing such observations 

 is that the scattering and absorbing power of the 

 atmosphere remains the same during the series of 



NO. 2771, VOL. I io] 



readings. The " solar constant " is arrived at by a 

 process of extrapolation. In an earlier bulletin Mr. 

 Knox-Shaw has directed attention to doubts as to the 

 validity of this assumption, and has shown that if the 

 air becomes less clear as the day progresses then a 

 negative correlation between the computed " solar 

 constant " and the computed transmission coefficient 

 is to be expected. 



It is now found that there is such correlation not 

 only at Helwan, but also at other places for which 

 observations have been published. The correlation 

 coefficient averages about o-6, and the value of the 

 determinations of the solar constant by the extra- 

 polation method is therefore much discredited. 

 Further evidence of its unsuitability is furnished by 

 the lack of correlation between the values of the solar 

 constant found at different stations on the same day. 

 It will be for the Smithsonian Institution to show that 

 destructive criticism on the same lines will not affect 

 the spectrobolometer observations on which the 

 evidence for the day-to-day variations of the " solar 

 constant " depends. 



In the year 1919 Prof. Abbot developed a new 

 method of observation based on the well-known fact 

 that the more the sunlight is obstructed by dust, etc., 

 the greater will be the glare surrounding the sun. 

 The question has been investigated by the use of a 

 " pyranometer " (fire-above-measure), as the instru- 

 ment for determining the strength of radiation from 



