340 



SCIENCE. 



[Vol. XXI. No. 542 



tions than for the same current in stronger solutions, up to 30 per 

 cent. This shows itself especially with thin plates, and also in 

 the shorter time required for thick plates to reach a maximum 

 polarization with weak currents. The greater change in tem- 

 perature and the greater change in concentration of weak solu- 

 tions may account for this. 



For currents between 0.1 and 0.2 ampere, the polarization on 

 the end electrodes was: — 



For Hj,SOi, 1.84; 

 " NaCI, 1.98; 



" CuSO^, 0.00, with Cu electrodes, though, if the current den- 

 sity was too great or the time long, the anode would oxidize and be- 

 come irregular. C. Fromme, in a paper, "Ueber das .Maximum der 

 galvanischen Polarisation von Platinelektroden in Schwefelsaure " 

 (Annalen d. Physik u. Cliemie, Band XXXIII., s. 80-126), states 

 that the maximum polarization -varies both with the concentration 

 and the relative size of the electrodes, the extreme limits being 

 given as 1.45 to 4.31 volts — the minimum polarization coinciding 

 with maximum conductivity. His method for measuring polari- 

 zation was somewhat similar to that used in this work. As bear- 

 ing upon "the change of polarization with time," I would refer 

 especially to the investigation of Dr. E. Root upon this subject, 

 discussed by Professor von Helmholtz, Wisch. Abh,, Vol.1., page 

 835. These experiments by Dr. Root seem to prove clearly that 

 the liberated ions penetrate deeply into the electrode, even when 

 liberated upon but one side of it, as in this case. I take great 

 pleasure in expressing here my thanks and deep obligation to 

 Professor A. Kundt and Dr. L. Arons for their kind sympathy 

 and direction in this work. 



Using CuSOj on one side of the partition, and H2SO4 on the 

 other side, careful determinations have developed the curious fact 

 that, although there is no visible development of ions (neither Cu 

 nor O) at the gold-leaf partition, yet the Cu does not pass 

 through the gold-leaf with the current, but H appears on the 

 cathode instead, provided the current density at the partition be 

 not greater than about .2 ampere per square centimetre. 



The " critical current-density" at which the ions just begin to 

 appear visibly on a gold-leaf partition varies for different electro- 

 lytes between the limit of 5.7 amperes for 30 per cent HsSOi and 

 sensibly zero for lead acetate. 



This " critical current-density " is proportioned to the conduc- 

 tivity of the electrolyte. It therefore also has a decided positive 

 temperature co-efBcient. 



ON THE FORMATION OF ALUMINUM SULPHATE IN THE 

 SHALES THROWN FROM COAL-MINES. 



BY M. H. LOCKWOOD, ASSISTANT IN THE DEPARTMENT OF GEOLOGT 

 AND MINEEALOQT, MISSOURI STATE UNIVERSITY. 



My attention was recently called to a white crystaline forma- 

 tion found on and between the layers of a red- colored shale that 

 is much used for walks in Columbia, Mo., and is obtained from 

 the old waste heaps of coal-mines in the vicinity. Upon exam- 

 ination I found it to consist of aluminum sulphate, which is 

 readily soluble in water, and has an alum-like taste. Occasion- 

 ally some iron sulphate is present. The question arose as to how 

 the aluminum sulphate was formed in between, and on, the lay- 

 ers of the shale. 



For the purpose of studying the formation, I visited the Reece 

 mine at Henry Station, oa the Wabash railroad, and there collect- 

 ed the following waste materials as thrown from the mine, viz., 

 fire-clay taken from below the^coal, clay-parting from a layer 

 about six inches from the bottom of the coal seam, iron pyrites 

 mixed with coal from spots throughout the coal seam, clay con- 

 taining iron pyrites and carbonaceous matter from just above the 

 coal, and a blue argillaceous shale from above the coal. 



The waste materials thrown from the mine, and exposed to 

 the air and moisture, go through the process of slacking or burn- 

 ing, and it is during this process that the aluminum sulphate is 

 formed. I also collected specimens from the burned and from 

 the burning heaps about the mine. 



Upon examination of the fresh specimens I found that the fire- 

 clay contained no free aluminum compound that would form 



aluminum sulphate after the slacking or oxidation of the heaps. 

 The clay-parting and other specimens containing iron pyrites and 

 carbonaceous matter, will oxidize so rapidly when exposed to the 

 air that the mass takes fire and we have iron sulphate and sul- 

 phuric acid formed. The sulphuric acid combines with the 

 aluminum in the shales and clays about it, forming aluminum 

 sulphate which crystalizes on the surface. 



The shale from above the coal contains some simple compound 

 of aluminum (probably the hydrate), and a considerable quantity 

 of free sulphur. The presence of the aluminum was shown by 

 the cobalt-nitrate test, and, also, when some of the shale was 

 boiled with hydrochloric acid and filtered, thesolutiongavea white 

 precipitate of aluminum hydrate upon the addition of ammonium 

 hydrate. 



Some pieces of the shale contained so much free sulphur that 

 they would burn, upon ignition, with a blue flame, giving off 

 fumes of sulphur dioxide. When some of the powdered shale 

 was leached with carbon dioxide, and the solution evaporated, a 

 residue of sulphur was obtained. These tests indicate that the 

 sulphur and aluminum thoroughly penetrate the shale. When the 

 heaps burn the sulphur becomes highly oxidized, and combines 

 with the aluminum, forming aluminum sulphate within the shale. 

 Heat drives the aluminum sulphate to the surfaces, hence it will 

 crystalize between the layers and on the surfaces of the shale. 



Free sulphur is found deposited in a crust at the top of the^ 

 burning heaps. This shows that there is an excess of free sul- 

 phur in the waste materials. 



The red color of the shale is due to the red oxide of iron formed 

 when the water is driven off by the heat. 



CURRENT NOTES ON ANTHROPOLOGY. — XXX. 



[Edited by D. O. Brinton, M.D., LL.D.. D.Sc] 



Prehistoric Ethnography of Northeastern Africa. 



There are two very learned and suggestive articles in the 

 Beitrage zur Assyriologie, Bd. II., Heft IF., 1892, which may be 

 combined to present the latest substantial opinions on the rela- 

 tions and sequence of linquistic stocks in the valley of the Nile 

 and the lands adjacent. The one is by Franz Pastorius, on the 

 Hamitic languages of East Africa; the other on the relations of 

 the Semitic and Old EgJ'ptian languages, by Fritz Hommel. In 

 what I present on the latter theme, I have also had the advan- 

 tage of a paper read before the Oriental Club of Philadelphia, by 

 the able egyptologist, Professor W. Max Miiller. 



Scarcely any question in early ethnography could be more im- 

 portant. It touches directly on the origin of the two oldest eiv- 

 ilizations of the world, — the Egyptian and the Babylonian. 

 According to Hommel, the Old Egyptian of the Pyramid Texts, 

 and the Old Babylonian (Semitic) tongues agree so closely in 

 grammar, in sequence of words, in phonetics, and in lexicog- 

 raphy, that their near relationship or their common origin must 

 be admitted. Professor Miiller informs me that in the Egyptian 

 of the Rammesside epoch at least sixty per cent of the words in 

 use were clearly Semitic. These relations are, however, dis- 

 tinctly not with the western Semites, but directly between the 

 eastern Semitic (Babylonian) and the Old Egyptian. Hommel 

 vei-y pertinently adds that this by no means justifies the conclu- 

 sion that the original home, die urspriingliche Heimat, of the 

 common stock was in the valley of the Euphrates; it might just 

 as well have been on the Nile. 



Some strength is given to the latter possibility by his compari- 

 sons of the Old Egyptian with the Berber dialects. He finds 

 that the lexicon of these latter is Old Lybian, but that their 

 grammar and syntax are very closely related to the Old Egyptian. 

 There is no doubt but that the ctiaracteristic forms of the perfect 

 and imperfect tenses were at one time common to the Berber, the 

 Old Egyptian and the Semitic tongues. Besides these, as pointed 

 out by Pastorius, the Hamitic (or Berber) dialects had in com- 

 mon with the original Semitic the personal pronouns, the femi- 

 nine in t, and a number of minor structural elements. He is 

 convinced that the East African Hamites (sometimes called 

 Kushites) have been dwellers on the upper tributaries of the 

 Nile, in Abyssinia, for many thousand years. Of their dialects, 



