May 20, 1875] 



NATURE 



51 



flower and bring their pollen -covered under-side first in 

 contact with the stigma, which slightly overtops the 

 anthers. The flowers of Lilium Martagon must conse- 

 quently be considered as adapted to cross-fertilisation by 

 Lepidoptera. 



The colour of these flowers, dark reddish brown, with 

 dark purple dots on the inside, is not very striking, and 

 in the daytime they are but slightly scented, whereas 

 during the evening they emit a very attractive sweet 

 odour. Hence we may safely conclude that they are far 

 more attractive to crepuscular and nocturnal than to 

 diurnal Lepidoptera. 



Thus far, in Thuringia, in July 1873, I had succeeded in 

 explaining the separate pecuUarities of the flowers ; but 

 in vain had I watched them repeatedly during the evening 

 in order to surprise the fertilisers in the very act of 

 fertilisation. But the hope I had failed in when making 

 every effort to realise it, happened to be fulfilled a year 

 later, quite unexpectedly. In the Vosges, returning from 

 the Hoheneck, and passing the village Metzerall, July 5, 

 1874, towards the evening, I was struck with the sight of 

 flowering plants of Lilium Marta^ott growing in a 

 garden hard by, and a specimen of Macroglossa stella- 

 tartim flying round them and fertilising them. 



Fig. 64.— a single 



magnified. 



Freely fixed in the air by the rapid movement of his 

 wings, this busy Sphinx inserted his long slender pro- 

 boscis into the honey-channels of the sepals and petals, 

 now of a single one, now of others of the same flower, and 

 having done so immediately flew away to another flower. 

 Yet, the flowers never being turned directly downwards, 

 but somewhat inclined, all the honey- channels of any 

 flower were never sucked by the Sphinx, but in every case 

 only those of the uppermost sepals and petals. When 

 sucking he always touched the stigmas and the anthers 

 with his legs and under-side, and the latter ones were to 

 be seen rocking and swinging. Thus, undoubtedly, the 

 under-side of the Sphinx was dusted with pollen, and the 

 stigma of the flower next visited, when first touched by 

 the pollen-covered under-side, was cross-fertilised. A single 

 Sphinx, with his vehement movements during a quarter 

 of an hour, may easily visit and cross-fertilise plenty of 

 flowers of Lilitivi Martagon. Nevertheless, self-fertili- 

 sation in many of these flowers will occur, where visits of 

 Sphingidse are wanting. For the stigma, by being bent 

 upwards more decidedly than the anthers, comes fre- 

 quently into contact with one or two of them ; and 

 C. Sprengel, who enclosed the yet unopened flowers of 

 L. Martagon in a net, thus excluding all insects except 

 some ants (and perhaps Thrips), was surprised to find that 

 every capsule developed and matured its seeds. 



Lippstadt Hermann Muller 



NOTE ON THE HYRCANIAN SEA 



THE resolution of the problems which are involved in 

 the physical aspects of Western Turkestan, and 

 which have offered so ample a scope for speculatioii, will 

 probably be one of the earliest and most important con- 

 sequences of the occupation of the banks of the Amii 

 Darya by Russia. But, whatever may be the light which 

 will thus be afforded to geographers, ethnologists, or his- 

 torians, it is to be expected that the field of mquiry will 

 widen and recede, in proportion as each step forward is 



made, along paths which have hitherto been shrouded in 

 obscurity. 



Among the observations which will demand, and which 

 will most certainly fully repay, the greatest attention, are 

 those which shall accurately determine the true rate of 

 evaporation from, the surface of Lake Aral. A meteoro- 

 logical observatory was established in June 1874 on the 

 lower courses of the Amu, and its working will contribute 

 much to a knowledge of the rate of local evaporation. It 

 may be doubted, however, whether such observations as 

 are recorded at Niikiis will be of practical value for deter- 

 mining the desiccation going on in Lake Aral itself. In 

 the absence of precise information we shall for some 

 years be dependent upon data of doubtful trustworthiness, 

 in regard to the aspect the lake may have presented at 

 different epochs in past history. 



Among such data there is an isolated observation 

 which seems worthy of more attention than has hitherto 

 been given to it. Between the years 1848 and 1858 

 Boutakoff found that the depth of water at the entrance 

 of Abougir (the gulf at the south-west corner of Lake 

 Aral, which is now entirely dry) had decreased by eighteen 

 inches, or, in other words, at the rate of 0*05 yards per 

 annum. This rate of decrease may possibly be not very 

 exact ; but it is approximately so, and may therefore 

 serve, until better data are available, to draw some con- 

 clusions regarding the Aralo-Caspian Sea. 



The chart of Lake Aral, compiled from the surveys of 

 1848-49, shows the waterspread to be about 24,500 square 

 miles. The contour Une drawn at a depth of twenty-four 

 feet on this chart includes an area of about 18,300 square 

 miles, i.e. the loss of surface is 6,200 square miles. For 

 every yard of fall below its surface of 1848, Lake Aral, 

 down to a depth of eight yards, loses a waterspread of 

 775 square miles. And since during the past twenty- 

 seven years the surface has fallen 27 X 0-05 = 1*35 yards, 

 the waterspread of 1875 will be 24500 - 775 X i'35 = 

 24500 — 1046*25 = 2345375 = 23454 square miles, say. 

 The mean of the two waterspreads of 1848 and 1875 

 will be ?_45^2L±13454 _ 47954 _ ,3977 square miles, or 



2 2 



74,271,155,200 square yards ; and this quantity multipUed 

 by 0*05 gives 3,713,557,760 cubic yards as the volume of 

 water lost by Lake Aral yearly since 1848, or a loss of 

 120 cubic yards per second. 



The supply poured into Lake Aral by the Amii and by 

 the Syr can only be guessed at, since it has probably 

 fluctuated during the past twenty-seven years. At the 

 present time the combined volume afforded by those two 

 rivers may be taken at about 2,000 cubic yards per 

 second ; and this estimate is probably not ten per cent, 

 removed from the actual truth. The evaporation, then, 

 from the lake must be assumed to have been, since 1848, 

 2000 -f 120 = 2120 cubic yards per second, from a water- 

 spread of 23,977 square miles, or 74,271,155,200 square 

 yards, which is equal to an evaporation of 00026 yards 

 per diem = 0*0936 inches per diem, or thirty-four inches 

 per annum. 



The physical aspects of the shores of Lake Aral suffice 

 to show that in very recent times its level has been at 

 least fifty feet higher than that of to day. With this 

 increased depth the waterspread would be about 36,500 

 square miles, or 1 13,062,400,000 square yards. The daily 

 evaporation from this surface at 0*0026 yards will be 

 293,962,240 cubic yards, or 3,400 cubic yards per second. 

 There was therefore a time (and that a recent one) when 

 Lake Aral received a supply of 3,400 cubic yards per 

 second ; and, indeed, of more than that quantity. The 

 Russian knowledge of the country, handed down by the 

 great map of the sixteenth century, informs us that a 

 river flowed from the Aral to the Caspian. The geogra- 

 phical MS. of (according to M. Vdmbdry) Ibn Said el 

 Belkhi, notices in the early part of the tenth century, the 

 opinion that the two seas communicated ; and this com- 



