May II, i! 



SCIENCE. 



225 



in atomic volume ; and hence the spectrum of H CI, for example 

 consists simply of the spectrum of hydrogen combined with 

 that of chlorine, with certain changes in the intensity of some 

 lines. 



In comparing the spectra of hydrogen and water-vapor, it was 

 found that wave-lengths of the lines of the so-called II* or com- 

 pound line-spectrum of hydrogen, which has been investigated by 

 Hasselberg, were twice those of the corresponding lines in the 

 water-vapor spectrum. This conclusion was arrived at by compar- 

 ing with the comparatively few lines of the water-spectrum that 

 were accessible at the time. To test the conclusion, however, a 

 list of wave-lengths that should be in the water spectrum was 

 drawn up and sent to Professor Liveing at Cambridge, and the 

 wave-lengths compared with those obtained by Liveing and Dewar 

 in their recent experiments. As the result of this comparison, the 

 author publishes a list of nearly sixty lines between wave-length 

 2800 and 2450, in which to each estimated line there corresponds 

 an observed one ; the difference between the observed and calcu- 

 lated wave-lengths in no case amounting to more than one Ang- 

 strom unit, or i part in 2,500. 



The author therefore concludes, on the basis of his theory, that 

 hydrogen, in that condition in which it gives this second or com- 

 pound line-spectrum, occupies twice the atomic volume which it 

 has in water-vapor. 



The primary or elementary line-spectrum of hydrogen, however, it 

 was found might be divided into two groups of lines, in such a manner 

 that the wave-lengths of the one group when multiplied by ij, and of 

 the other when multiplied by |, gave the wave-lengths of the corre- 

 sponding lines in the H,0 spectrum. Whence the author, by means 

 of his fundamental theorem, reasons thus : hydrogen is formed of 

 two primary elements, which may be designated a and b, and 

 which give rise to the two parts of the elementary hydrogen spec- 

 trum under each other's influence. Let a and b represent the vol- 

 umes of these two substances respectively in unit-volume of hydro- 

 gen ; then « 4- ^ = i ; and, since hydrogen occupies two-thirds the 

 atomic volume in water-vapor that it does in the primary condition, 

 from the fundamental theorem we have 



M« + 1*^ = f • 

 From these two equations, 



fl = I, (5 = i ; 

 therefore hydrogen is a combination of the form H = ba, and is thus 

 analogous to ammonium (N H4), and, as Professor Griinwald as- 

 serts, will, on dissociation, expand in the ratio of 3 to 2. 



The primary element a must be a gas many times lighter than 

 hydrogen. The spectra of these two elements, a and b, in the free 

 condition may be at once obtained from the hydrogen spectrum by 

 the previous theorem, when it is granted that the gas, on dissocia- 

 tion, expands in the ratio of 3 to 2 ; for we have only to multiply 

 the wave-lengths of the group a in the hydrogen spectrum by | to 

 obtain those of the substance a in the free condition ; and in like 

 manner the wave-lengths of the substance b may be obtained from 

 the corresponding group b. 



Professor Griinwald has tabulated five lines in the spectrum of a 

 between wave-lengths 9842 and 5653, and about forty lines of 

 the spectrum of b, and each is found to correspond with a line in 

 the solar spectrum. He concludes, therefore, that hydrogen, in the 

 dissociated condition, exists in the sun, and identifies one of the 

 lines of b with the so-called Helium line," 5874.9 of Angstrom's 

 scale, while reasons are given for believing the corona line (1,474 

 of Kirchoff's map) is one of those in the spectrum of a. These 

 two component elements of hydrogen he therefore suggests might 

 be named ' Coronium ' and ' Helium.' 



From similar considerations to those employed in the case of hy- 

 drogen, oxygen, in its simplest molecular condition, is found to con- 

 sist of the modified hydrogen, which gives the secondary spectrum 

 before mentioned, with an equal volume of a substance O', with 

 which it combines without change of volume. This O' is a combi- 

 nation of four parts by volume of the same element {b) which was 

 found in hydrogen, with five parts of another substance (O"), which 

 is itself composed of four parts of b with five parts of an unknown 

 primary substance c. The formula expressing the above is, 

 O = H' [b,ib^c-:i,-\. 



In a long paper published in December in the Sitzungsberichte 



der Kazs. Akad. der Wissenschafteii of Bohemia, Professor Griin- 

 wald has extended his work to the spectra of magnesium and carbon, 

 employing the wav-e-lengths as determined by Liveing and Dewar, 

 and Hartley and Adeney, with the result that the spectrum of 

 magnesium may be separated into four groups. The first is due to 

 ' Helium,' neither ' condensed nor dilated ; ' the second is that of the 

 primary element c in the condition in which it exists in oxygen ; the 

 third is that of b in the state in which it exists in free hydrogen ; 

 while the fourth is caused by the same element b, but in the chem- 

 ically more ' condensed ' state in which it exists in water-vapor. 



There are still a number of weak magnesium lines which fall 

 naturally into these groups, but the corresponding lines to which in 

 the hydrogen and oxygen spectra are not known to exist. Carbon 

 has similarly been resolved into a certain compound of these ele- 

 ments /; and c. 



These speculations will require most thorough investigation and 

 testing before they can be accepted ; but the first point to be seri- 

 ously examined is the basis on which they rest. If the coincidences 

 reported by Professor Griinwald, when examined carefully, are found 

 sufficiently close and numerous to prove that a large group of lines 

 in the spectrum of one substance can be obtained by simple multi- 

 plication by a constant multiplier from a corresponding group in 

 the spectrum of another substance, and if there is any other fact, 

 such as the regular periodic arrangement of the lines, which would 

 seem to connect that group of lines together, then it is one of the 

 most important facts which have yet been developed in connection 

 with spectra. But it is necessary that the agreement should be of 

 the same order of accuracy as the errors in the determination of 

 wave-lengths, and there should appear some other fact connecting 

 a group of lines together. As to the 'condensation ' theory, noth- 

 ing need be said until the facts are more thoroughly worked up ; 

 but the remark of its author, that the intensity of the lines due to a 

 substance will experience great differences in intensity in different 

 combinations, while undoubtedly true, gives great elasticity to the 

 theory, and admits of its adaptation to so wide a range of facts as 

 to seriously weaken the evidence advanced in its favor. 



HEALTH MATTERS. 

 Diphtheria in New York. 



The prevalence of diphtheria in New York and Brooklyn has 

 awakened a renewed interest in the means for its prevention. A 

 paper on this subject was recently read by Dr. A. Caille before^ the 

 New York Academy of Medicine, and is reported in the Neiv York 

 Medical Journal. It had been his experience, as it probably had 

 that of many other physicians, that in certain families one or more 

 members regularly had diphtheria in the spring or autumn. This 

 was particularly true of children. It had occurred to him that such 

 persons might harbor the microbes, or other essentials to the de- 

 velopment of the disease, in the nasal and oral cavities. The germs 

 of diphtheria would readily take hold of damaged mucous mem- 

 brane. 



In trying to establish the correctness or fallacy of this view of 

 self-infection, he had selected eight cases, in all of which the 

 patients had suffered from true diphtheria twice or more prior to 

 October, 1885. The families were well known to him, and they had 

 occupied the same houses or had the same surroundings for a num- 

 ber of years. The parents of the children were intelligent enough 

 to carry out his instructions. All carious teeth were to be filled or 

 extracted, the teeth to be examined from time to time ; the mouth 

 was to be thoroughly rinsed three times a day, after each meal, 

 with either a three-per-cent solution of chlorate of potassium in 

 water, a five-per-cent solution of liquor sodir chlorates, or a satu- 

 rated solution of borax in water. Besides using it as a mouth wash 

 and gargle, some of the solution was to be drawn into the nose. 

 From October, 1885, to December, 1887, not one of the persons 

 experimented upon suffered from diphtheria, although five of them 

 had several attacks of acute pharyngitis and amygdalitis. There 

 was diphtheria in the family of three of the number, but they did 

 not contract the disease. While these cases were insufficient to 

 furnish absolute proof of the benefit of such prophylactic measures, 

 yet they went far to establish the belief, that, if the nasal and oral 



