July 21, 1892] 



NATURE 



287 



P'ree azoimide is fiaally obtained by distilling the crystals of 

 the sodium salt with dilute sulphuric acid, and repeatedly 

 re-distilling over fused calcium chloride the hydrated liquid 

 which firit passes over. 



As an alternative method which has some advantages as 

 regards facility of manipulation, Prof. Curtius employs the 

 hippuryl derivative of hydrazine instead of the benzoyl com- 

 pound The product of the action of nitrous acid upon this 

 compound is a substance which can be readily isolated in 

 crystals, and if these crystals are dissolved in dilute caustic 

 soda, the solution at once yields azoimide upon distillation 

 with dilute sulphuric acid. 



Before describing the inorganic synthesis of Prof. Wislicenus, 

 it may be mentioned that a still simpler organic synthesis of 

 azoimide from the long known diazobenzene imide, CbHjNj, 

 has been achieved by Drs. Noelting and Grandmougin. Al- 

 though diazobenzene imide itself is too stable a substance to 

 yield azoimide directly by simple saponification with soda, these 

 chemists found that its dinitro derivative yielded directly to the 

 attack of an alcoholic solution of potash, the potassium salt of 

 azoimide bein» formed, which of course gave free azoimide upon 

 distillation with dilute sulphuric acid. 



The inorganic synthesis of azoimide now achieved by Prof. 

 Wislicenus depends upon the interaction of ammonia gas and 

 nitrous oxide in the presence of heated metallic sodium. Am- 

 monia and nitrous oxide do not act directly upon each other, 

 not even when a mixture of the two gases is passed over caustic 

 bases — soda- lime for instance. But they react readily in 

 presence of metallic sodium. The explanation of this lies in 

 the fact that the sodium amide discovered by Gay Lussac and 

 Thenard is first formed, and this compound reacts with the 

 nitrous oxide with production of the sodium salt of azoimide : — 

 NaNHo + NoO = NaNj + H2O. 

 The water produced at the same time reacts with one-half of the 

 sodamide, forming caustic soda and liberating ammonia gas : — 



NaNHo -f- HjO = NaOH 4- NH3. 

 Hence the complete reaction may be expressed by the equation — 

 2NaNHo -t- NoO = NaNg + NaOH + NH3. 

 As the sodium salt is less explosive than most of the other 

 salts of azoimide, requiring a higher temperature and not being 

 sensitive to percussion, the experiment is not dangerous if proper 

 care is exercised, and even if local explosions do occur they have 

 not yet been observed to shatter the glass tube. Unfortunately 

 glass is somewhat strongly attacked during the reaction, but if 

 iron tubes are employed the reaction is not so completely under 

 control. 



In actually conducting the experiment, metallic sodium, in 

 pieces not exceeding half a gram in weight, is placed in a series 

 of large porcelain boats, which are then laid in a glass combus- 

 tion tube, from which the air is subsequently displaced by means 

 of a current of ammonia gas. The tube is heated carefully in a 

 combustion furnace, when the sodium fuses and gradually passes 

 into sodamide. When all the metal has been thus changed, the 

 stream of ammonia is replaced by one of dry nitrous oxide. The 

 temperature should now be lowered to between 150° and 250^, 

 and for this purpose Prof. Wislicenus surrounds it by an iron 

 explosion chamber, which forms a capital air-bath, the tempera- 

 ture of which can be regulated by observing a thermometer or 

 thermometers inserted in it. The sodamide now slowly in- 

 creases in bulk and becomes converted into the sodium salt of 

 azoimide. As soon as ammonia ceases to be carried away in 

 the stream of issuing nitrous oxide the reaction is completed. 

 Upon cooling the sodium salt is found as a porous pumice-like 

 substance, much distended by the escaping ammonia. 



The sodium salt of azoimide is also formed when ammonia and 

 nitrous oxide gases are simultaneously passed over melted 

 sodium ; the yield, however, is not so large, and there is danger 

 of the sodium inflaming in the nitrous oxide. 



The fact that the sodium compound obtained is the sodium 

 salt of azoimide has been proved both by direct analysis (a 

 determination of nitrogen yielding close upon the theoretical 

 amount) and by its properties. The product of the reaction on 

 being removed from the combustion tube was thrown into water, 

 and the filtered solution distilled with dilute sulphuric acid. 

 The distillate possessed the intolerable odour characteristic of 

 azoimide, and behaved exactly like a solution of that substance 

 in water. It gave precipitates with nitrates of silver, mercurous 

 mercury and lead, which when separated and dried were found 

 to possess all the properties of the silver, mercury, and lead salts 



NO. \ 186, VOL. 46] 



of azoimide respectively. The fact that these salts were those of 

 azoimide was indeed sufficiently apparent from their violently 

 explosive nature, and the characteristic flames which were pro- 

 duced during their explosion. Moreover, gold dust was rapidly 

 dissolved with production of the red solution described by Prof. 

 Curtius, 



A quantity of the silver salt was subjected to analysis, and 

 was found to contain 717 per cent, of silver, the amount calcu- 

 lated for AgNa, being 71-8. 



Instead of sodium, either potassium or zinc may be em- 

 ployed. Potassium answers almost as well as sodium, forming 

 first an amide when heated in a current of ammonia, which is 

 subsequently converted by nitrous oxide into the potassium salt 

 of azoimide. Zinc likewise behaves in a similar manner, but the 

 yield of the zinc salt of azoimide, ZnN^, is not so good as in the 

 cases of sodium and potassium. To a greater or less extent, 

 therefore, it would appear that metallic amides when heated in 

 a current of nitrous oxide are generally converted into salts of 

 azoimide. The alkali metals, however, appear to be best suited 

 for practical use. A. E. TuxTON. 



THE REPORTED VOLCANIC ERUPTION AT 



GREAT SANGIR. 

 ACCORDING to a Renter's telegram from Sydney,^ de- 

 "^^ spatched on July 17, the vessel Catterthun, belonging to 

 the Eastern and Australasian Steamship Company, which had 

 arrived at Sydney from China, brought a report of a terrible 

 disaster in the vicinity of the Philippine Islands. She called on 

 her voyage at one of the chief ports of the island of Timor, 

 where rumours had been received according to which the island 

 of Sangir, situated between Celebes and Mindanao, had been 

 destroyed by a volcanic eruption. The whole population, num- 

 bering 12,000, was reported to have perished. The captain of 

 the Catterthun stated that on the voyage his vessel passed 

 through some miles of volcanic debris. 



We may not for some time receive further details as to the 

 real extent of the disaster reported by the captain of the Catter- 

 thun, but in the meantime the following account, by Mr. Sydney 

 J. Hickson, author of "A Naturalist in North Celebes," of the 

 island and of the history of its volcanic energy — which appeared 

 in the Titnes of Tuesday — will be read with interest : — 



Sangir, or "Great Sangir," as it is more frequently called 

 by the natives of the Archipelago, is the largest of a chain of 

 volcanic islands that connects the northern peninsula of Celebes 

 with the southern point of the island of Mindanao. The 

 islands, rising abruptly from the floor of the very deep Celebes 

 sea — a depth of over 2,000 fathoms was found by Her Majesty's 

 ship Challenger quite close to Great Sangir — are very mountain- 

 ous and covered by dense tropical forests. 



The islands Ruang and Siauw are both little more than vol- 

 canoes standing in the sea, but Sangir is a large island 25 miles 

 long by about 15 miles broad, with undulating hills and valleys 

 occupying its southern half, and the great Awu volcano and its 

 slopes the greater part of its northern half. 



When I visited the islands in November, 1885, the Ruang 

 and the Awu were quiet, but the Siauw was sending out dense 

 volumes of smoke that varied little in intensity from day to day. 

 From the accounts I received from the natives and from the 

 records of the islands in the Dutch books of travel, it seems 

 that the Siauw volcano has never been very violently active, but 

 both the Ruang and the Awu have a history full of most terrible 

 and heart-rending episodes. Of the Ruang I need not say 

 much. The last serious eruption occurred in 1871, when at 

 least 400 persons lost their lives either by the sudden rise of the 

 sea water that accompanied the eruption, or by the showers of 

 stones and ash. Of the Awu volcano we find recorded in 

 Valentijn's "Oud en Nieuw Oost Indien" that a most terrible 

 eruption occurred which lasted from the loth to the i6th of 

 December, 171 1. Sjamsialam and his son, the Princess Lorolabo 

 and her daughter Sarabanong, and over 2000 people of the king- 

 dom of Kandahar were killed. On March 2, 1856, there was 

 another fearful eruption, which lasted until March 17, and de- 

 stroyed nearly 3000 human lives. The streams of boiling 

 water and of steam which poured down the mountain slopes 

 rather than the flow of lava caused the enormous mortality of 

 this second eruption. After the eruption of 171 1 it seems that 

 a large lake of water was formed in the crater, and a certain privi- 

 leged class of Sangirese were allowed by the gods to visit this lake 

 every three or four months to test the water with their rice. If 



