Dec. 15, 1887] 



NATURE 



165 



rations, and the results I have obtained may be summed up as 

 follows : — 



Both the male and female sexual cells arise in the ectoderm of 

 the ccjcn Ksarcal canal system. At an early stage they perforate 

 the mes )glica and take up a position in the endoderm. 



The ova at an early stage become stalkeJ. The stalk of the 

 ovum, which is simply a modified pseudopodium, serves to keep 

 the ovum attached to the mesogloea. The stalk is sometimes 

 completely withdrawn, and the ovum' by amoeboid movements 

 migrates along the lumen of the canals to a more favourable 

 locality. 



Maturation and impregnation occur while the ovum is still in 

 the canals. 



The mature ovum is very small (l/loo mm. in diameter), and 

 is alecilhal ; nevertheless, it does not segment. 



The germinal vesicle of the fertilized ovum splits up into a 

 number of fragments, which, after a curious series of movements 

 in the ovum, are eventually scattered over its substance. 



By the time these fragments are thus .scattered over the ovum, 

 they have reached a considerable size, and, from faint markings in 

 the substance of the ovum, no doubt can be retained that they 

 are in reality the true nuclei of a morula stage in the develop- 

 ment of the embryo. The embryo next assumes the form of a 

 solid blastosphere, and its subsequent history is lost. 



It will be a very interesting point to determine the precise 

 mode of discharge of the embryo. I am very strongly of 

 opinion that the embryo is discharged by the mouth of the 

 gastrozooid, but I was, of course, unable to observe this in the 

 living state. Whether this is correct or not, the fact remains 

 that I have been unable to find in any of my preparations any 

 trace of a free or fixed gonophore, containing either embryos 

 or ova. 



In the development of the spermatoza, a similar phenomenon 

 is found to that in the development of the embryo. The 

 spermospore does not divide into a sperm-morula, the nucleus 

 alone fragments, and the subsequent formation of spermoblasts 

 does not occur until a very late stage. When the spermoblasts 

 are mature they are found in simple sporosacs on the dac- 

 tylozooids. The sporosacs exhibit no traces of any medusoid 

 structure. 



These researches tend to prove that the Milleporidae belong 

 to a separate stock of the Hydrozoa from the Hydromedusse, a 

 stock which probably never possessed free-swimming medusiform 

 gonophores. 



There seems to be no true relationship between Millepora and 

 Hydractinia. The absence of segmentation in the developing 

 embryo may probably be accounted for by the amoeboid move- 

 ment which it exhibits after development has commenced. The 

 evidence before us does not support the view that the ovum 

 of Millepora formerly contained much yolk, and has subsequently 

 lost it. 



Physical Society, November 26. — Dr. Balfour Stewart, 

 President, in the chair. — Mr. Asutosh Makhopadhyay was 

 elected a member of the Society. — The following communica- 

 tions were read :— On the analogies of influence-machines and 

 dynamos, by Prof. S. P. Thompson. The author pointed out 

 that in nearly all influence-machines there are two stationary 

 parts ("inductors ") electrified oppositely, which are analogous 

 to the field-magnet of dynamos, and a revolving part carrying 

 '"sectors" which correspond to the *'f^<r//c?«5" of an armature. 

 To prevent ambiguity Prof. Thompson proposes to call the in- 

 ductors ^^ field plates," and the revolving parts as a whole an 

 *' armature." In the Wim-.hurst machine both field plates and 

 armature rotate, and each acts as field plates and armature 

 alternately. In the two field plate influence-machines there are 

 four and sometimes six brushes. Two of these act as potential 

 equalizers, two as field plate exciters, and the remaining two (if 

 any) are generally placed in the " discharge " or external circuit. 

 The Hollz machine having only four brushes, two serve the 

 double purpose of potential equalizers and discharge circuit, and 

 this machine excites itself best when the discharging rods are in 

 contact. In this respect it resembles a series dynamo which only 

 excites itself when the external circuit is closed, but on opening 

 the circuit (say by inserting an arc lamp) produces remarkable 

 effects. So in the Holtz machine on separating the discharging 

 knobs a shower of sparks results. The Toepler machine (made by 

 Voss) having six brushes resembles a shunt dynamo, and excites 

 itself best on open external circuit. Analogies were traced 

 between Thomson's replenisher and the Griscom motor. Arma- 

 tures of influence-machines, as in dynamos, can be divided into 



ring, drum, disk, and pole armatures, and examples of each kind 

 were mentioned. The " Clark Gas Lighter" is a good example 

 of a drum armature, and a diagram showing the internal arrange- 

 ments was exhibited. An example of an analogue to the com- 

 pnind dynamo was mentioned as existing at Cambridge, in the 

 form of a Holtz machine believed to have been modified by Clerk 

 Maxwell. Another analogue with dynamos is found in the dis- 

 placement of the electric field when the armature is rotated, just 

 as the magnetic field of a dynamo is shifted round in the direc- 

 tion of rotation. Further analogies were traced between 

 "critical velocity " of dynamos (which depends on the resistances 

 in the circuit) below which they do not excite themselves, and a 

 similar critical velocity of influence-machines ; e.g. in a Wims- 

 hurst or Voss machine, the potential equalizing circuit should 

 have a low resistance if they are to excite themselves readily. 

 Self-exciting dynamos excite better when the iron is bad and 

 retains the magnetism, and influence-machines excite better when 

 the field plates are made of paper or such substance as can well 

 retain a residual charge. Finally an apparatus analogous to 

 Thomson's "water-dropping accumulator" was exhibited, in 

 which an electric current was generated by mercury falling down 

 a tube between the poles of a magnet. — 'On the eff'ect produced 

 on the thermo-electric properties of iron when under stress or strain 

 by raising the temperature to a bright red heat, by Mr. Herbert 

 Tomlinson. In June last the author described some remarkable 

 "effects of change of temperature on twisting and untwisting 

 wires which have suffered permanent torsion," of which the 

 present paper is a continuation. It is found that at or about 

 the critical temperature (a bright red heat) mentioned in the 

 previous paper, a sudden E.M.F. is generated at the junction of 

 two iron wires, one of which is under stress or has suffered 

 permanent strain, and the other in an unstrained state. By 

 suddenly bringing a red-hot iron wire in contact with cold iron, 

 an E.M.F. of about 1/20 volt is produced. If copper be used 

 the E.M.F. is about \ volt. The author also showed that if one 

 part of an annealed iron wire is heated to a bright red by a 

 bunsen flame, an E.M.F. is generated if the position of the flame 

 is slightly altered, the direction of the E.M.F. depending on the 

 direction of the displacement. Prof. Ayrton believed the high 

 E.M.F. exhibited by hot and cold copper was really due to 

 oxide of copper ; and Prof. S. P. Thompson said that different 

 effects could be produced by using the oxidizing or reducing 

 parts of the flame in heating the wire. — On the method of dis- 

 criminating real from accidental coincidences between the lines 

 of different spectra, with some applications, by Mr. E. T. J. 

 Love. 



December 10. — Prof. W. E. Ayrton, Vice-President, 

 in the chair. — Mr. E. A. C. Wilson, and Mr. W. E, 

 Sumpner were elected members of the Society. — Mr. H. 

 G. Madan described the optical properties of phenyl-thio- 

 carbimide. This body, derived from aniline, is a colourless 

 liquid, density i '35° C, and of high boiling-point 222" C. The re- 

 fractive indices for the A and G lines are 1*639 and 1*707 

 respectively. It is thus seen to be a highly refractive liquid, 

 and to have about the same dispersive power as carbon-bisulphide, 

 whilst its use in prisms is unattended by many of the risks and 

 inconveniences experienced with carbon-bisulphide. The dis- 

 persion at the blue end of the spectrum is very marked. Being 

 less mobile than carbon-bisulphide, it is less affected by con- 

 vection currents. The "refractive equivalent" calculated from 

 its chemical constitution differs considerably from the observed 

 value, and this difference the author believes due to the presence 

 of the phenyl radicle and sulphur atom. A polarizing prism 

 made on Jamin's plan, but using phenyl-thio-carbimide as the 

 liquid, gives a fairly wide angular field (about 25°). Mr. Hilger 

 stated that there was no great need of liquid prisms now, for 

 very dense flint glass could be obtained with mean index of 

 about I "8. Dr. Perkin has recently supplied him with Canada 

 balsam perfectly colourless, and which does not tarnish the 

 polished faces of spar ; hence one of the greatest objections to 

 the use of Canada balsam in spar polarizing prisms has been 

 removed. Dr. Gladstone pointed out that the constants for the 

 phenyl radicle and for sulphur atoms had been determined, and 

 thought the calculated " refractive equivalent " obtained by in- 

 cluding these would be much nearer the observed value than the 

 one given by Mr. Madan. — On the recalescence of iron, by 

 Mr. n. Tomlinson. If an iron bar which has suffered per- 

 manent strain be heated to a white heat and allowed to cool, 

 the brightness at first diminishes and then reglows (recalesces) 

 for a short interval. Under favourable circumstances as many 



