May 1 6, 1518] 



NATl/Jm 



217 



charge and the poisonous substance is necessary. Both 

 sides are busy trying to find something that the others 

 have not used, and both are trying to find a " colour- 

 less, odourless, and invisible " gas that is highly 

 poisonous. It is within the realm of possibilities that 

 the war will be finished, literally, in the chemical 

 laboratory. 



The following compounds have been used by the 

 Germans in gas-clouds or in shells : — 



1. Allvl-iso-thiocyanate (allyl mustard oil), CjHsNCS 

 (shell). ' 



2. Benzyl bromide, CH^-CH^Br (shell). . 



3. Bromo-acetone, CHjBr.CO.CH, (hand grenades). 



4. Bromated methylethyl-ketone (bromo-ketone), 

 CHoBr.CO.QH, or CH^.CO.CHBr.CH, (shell). Di- 

 bromo-ketone,. CH^.CO.CHBr.CH^Br (shell). 



5. Bromine, Brj (hand grenades). 



6. Chloro-acetone, CH2CI.CO.CH3 (hand grenades). 



7. Chlorine, Clj (cloud). 



8. Chloromethyl-chloroformate (palite), 



ClCOO.CH,.ei (shell).: 



9. Nitrotrichloromethane (chloropicrin or nltro- 

 chloroform), CCUNO^ (shell). 



10. Chlorosulphonic acid, SO3.H.CI (hand grenades, 

 and " smoke pots "), 



11. Dichlorodiethylsulphide (mustard gas), • 



(CHX1.CH3),S (shell). 



12. Dimethyl sulphate, (CH3)j'SOi (hand grenades). 



13. Diphenvlchloro-arsine, (C8H5),AsCl (shell). 



14. Dichloromethyl ether, (CH,Cl"),0 (shell). 



15. Methylchlorosulphonate, CH3CISO3 (hand gren- 

 ades). 



16. Phenylcarbylamine chloride, C.H^N.CClj (shell). 



17. Phosgene (carbonyl chloride), COClj (cloud and 

 shell). 



18. Sulphur trioxide, SO3 (hand grenades and shell). 



19. Trichloromethyl - chloroformate (diphosgene, 

 superpalite), CICOO.CCl, (shell). 



20. Xylyl bromide (tolyl bromide), CH3.CeH,.CH,Br 

 (shell). 



CRYSTALS OBTAINED FROM GLASS 

 FURNACES. 



TVyjR. G. V. WILSON has studied the materials 

 ■i'-* obtained from glass furnaces of the tank type 

 making bottle glass, where, by a rupture of the tank, 

 the glass has flowed out and has been allowed to 

 crystallise slowly, and he described his results to the 

 Society of Glass Technology on April 17. He finds 

 that crystals of wollastonite form in great numbers, 

 partly arranged in spherulitic groups of long diverging 

 crystalline fibres, partly as separate rod-shaped crystals 

 with well-defined faces. Tridymite occurs also in flat 

 hexagonal plates, very thin, but showing the polarisa- 

 tion in sectors which is so characteristic of this 

 mineral. He has also observed quartz as double hexa- 

 gonal pryamids in places where veins of glass have 

 eaten their way into the bricks which make the walls 

 of the tank ; and manganese augite, of purplish-brown 

 colour, only in parts of the glass where an excess of 

 manganese oxide has existed through imperfect mixing 

 of the ingredients of the batch. 



Fragments of limestone, probably in part magnesian, 

 occurred in the glass, imperfectly dissolved and showing 

 recrystallisation through contact alteration and admix- 

 ture with silica and other elements of the glass. These 

 have a granular crystalline structure except where 

 veins of glass penetrate inta them. The new minerals 

 produced are wollastonite, .augite (golden-yellow,^ in 

 small prisms), melilite, and prc>bablv a silicate of lime 

 (3CaO,2SiO,). ' ; • ■ • 



The vault of the furnace consists of firebrick, and 



NO. 2533, VOL. lOl] 



is covered with a fused glassy, layer, . from which 

 stalactites- hang down, and drops of molten matter 

 must have been -falling into the gkss below. The 

 zone of altered brick is about an inch thick, and two 

 layers can be detected in it — an outer glassy stalactitic, 

 layer containing much corundum and a little sillimanite. 

 in a glassy matrix, and an inner white layer looking 

 very like white porcelain to the naked eye. This 

 inner layer is richly cj?arged with sillimanite needles. . 



The external surface of glass pots also contains, 

 very well fqrmed crystals of sillimanite, and sometimes, 

 also magnetite and corundum. The inner surface of 

 old glass pots often shows much sillimanite embedded 

 in clear glass, and, where the cooling has been slow». 

 biotite an4 oligoclase also make their appearance. 



'Nlr...'y\yitson regards as important the presence of 

 volatile fluxes, such as soda, given off by the heated, 

 glass ^ These combine with the alumina arid silica of 

 the day, forming glassy alkali-alurriina-silicates which 

 are comparatively fusible, and serve as a medium in 

 which corundum, sillimanite, and other minerals are 

 crystallised. The corrosion of the glass pots is due 

 largely to the action of this alkali-alumina-silicate melt 

 eating its way deeper and deeper into the clay. 



THE CARNEGIE INSTITUTION O^ 

 WASHINGTON., 



THE Year Book for 19 17 of the Carnegie Institu- 

 tion of Washington has now reached us in its 

 complete form. Attention has already been directed to 

 the report of the president of the institution with 

 which the Year Book, which was published previously 

 separately, opens, and the opportunity provided by the 

 receipt of the complete volume may be taken to refer 

 to the financial records of the institution. During the 

 year ending October 31, 19 17, the total financial re- 

 ceipts of the institution reached 285,120^. ; of this' 

 amount i2o,iooZ. represented interest on endowment,, 

 2o,iooi. interest on deposits in banks, and 36,000!. 

 amounts derived from miscellaneous sources. The 

 total of the yearly incomes of the institution since its 

 foundation in 1902 amounts to 2,886,6651. 



The income of 1917 was expended in the manner 

 shown in the following table : — 



£ 



Investment in bonds 101,100 



Large projects 139,160 



Minor and special projects i9.,Soo 



Publications 12,600 



Administration 9,800 



Total 



282,160 



The departments of investigation to which the larger 

 grants were made by the trustees of the institution are 

 shown below : — 



£ 



Botanical research 8,686 



Embryology 6,954 



Experimental evolution 11.386 



Geophysical laboratory ^20,302 



Historical research 7,090 



Marine biology 3.980 



Meridian astronomy 6,231 



Nutrition laboratory 9,227 



Publications ... ... ... ■■• 2,280 



Solar observatory 35, 5^9 



Terrestrial magnetism 28,441 



Total 



140,086 



The publication of twenty-five volumes was author- 

 ised during the year at a cost of 13,000?. 



