May 26, 1892] 



NATURE 



93 



that which obtains in the case of the first root. The stipular 

 structures, so characteristic of the Marattiaceoe, are entirely 

 absent from the first two leaves, but appear in a well-developed 

 condition on the third and all succeeding leaves. 



" On the Shoulder-girdle in Ichthyosauria and Sauropterygia." 

 By J. W. Hulke, F.K.S. 



The author discusses the structure of the shoulder-girdle and 

 the homologies of its several parts in these families. He shows 

 that the alleged existence of a precoracoid in the Ichthyosauria 

 rests on an insufficient foundation ; otters proofs that in Plesio- 

 sauria the anterior ventral ray is not only theoretically but 

 actually precoracoid ; and also that the dorsal ray in the girdle 

 is homologous with the shoulder-blade in Testudinata and other 

 Reptilia. 



"On the Development of the Stigmata in Ascidians." By 

 Walter Garstang, M. A. 



The author shows that the transverse rows of stigmata, which 

 are present in all the fixed Ascidians, may arise in one or the 

 other of two ways : either as independent perforations, distinct 

 from the first (oozooid of Clavelina, buds of Botrylltts) ; or as 

 secondary formations, due to the subdivision of a series of pri- 

 mary transverse stigmata on each side (oozooid of Botrylltts, 

 Styelintr). The former method of development is shown to be 

 a modification of the latter. 



The primary transverse stigmata (or " protostigmata ") of 

 Botrylltis and the Styelina agree precisely with the definitive 

 stigmata of Pyrosoma in structure, position, and order of for- 

 mation. They are accordingly regarded as homologous 

 formations ; and the conclusion is drawn that Pyrosoma has not 

 been derived from the fixed Ascidians, but represents an 

 ancestral type of Caducichordate Tunicata, antecedent to the 

 whole of the phylum Ascidiacea. 



Physical Society, May 13.— Dr. E. Atkinson, Treasurer, in 

 the chair. — Mr. R. Inwards read a paper on an instrument 

 for drawing parabolas. It was designed for drawing curves of 

 short focus such as are required for reflectors and for diagrams 

 of the paths of comets and projectiles Its construction is based 

 on the fundamental property thU every point on the curve is 

 equidistant from the focus and the directrix. In the diagram 

 btlow, AB is a slot representing the directrix, and fghi 



a rhombus jointed at the corners and pivoted at F, whilst CD 

 indicates a bar capable of sliding through guides at H and I. 

 A rod, LE, is coupled to G by a bar, GM, such that the lengths 

 LM, ME, and MG are equal. As L and G slide along AB, the 

 point E describes a parabola whose latus rectum depends on the 

 distance of f from ab. In the instrument F is carried by 

 a slotted arm so that its position is adjustable. GE is always 

 perpendicular to ab and equal to EF. Prof. Boys inquired 

 whether the instrument could be modified to draw any conic 

 section by arranging that the ratio of EF to eg instead 

 of being unity, might be greater or less than unity. Prof. 

 Perry said an instrument for drawing curves represented by the 

 equation;/ = x" was greatly needed for engineering work. — Mr. 

 F. H. Nalder exhibited and described some electrical instru- 

 ments. The first shown was a ballistic galvanometer with one 

 pair of coils, the distinguishing features of which were accessi- 

 bility, small damping, great sensitiveness, and the arrangement 

 of the control. The control is eff-cted by a "tail magnet" 

 carried on a horizontal tube supporte^l by a pillar outside the 

 case, as suggested by Prof. R. M. Walmsley. A small magnet 

 on the cover serves (or zero adjustment. The suspended system 

 consists of four bell magnets, two being in the middle of the 

 coil and one at top and bottom respectively, arranged so as to be 

 astatic. The sensitiveness of the instrument shown was such 

 that i of a microcoulomb gave 300 divisions (fortieths of 

 an inch) when the periodic time was 10 seconds and scale 



NO. I I 78, VOL. 46] 



distance about 3 feet. Resistance of galvanometer about 

 lo.coo ohms. To bring the needle to re.st quickly, a damping 

 coil mounted on an adjustable stand, and a special reversing key 

 with resistances in its base, are provided. The key has succes- 

 sive contacts arranged so that when pressed lightly, only a weak 

 current passes round the damping coil, whilst when pressed 

 further a much stronger current passes. The strong currents are 

 used to check the large elongations, and the weak ones for finally 

 bringing to zero. A lamp stand with semi-transparent scale 

 arranged for use with a glow lamp was next shown. Instead of 

 reading by the image of the filament, as is ordinarily done, the 

 lantern is arranged to give a bright disk of light with a black line 

 across the middle. Mr. Blakes^ey asked if the galvanometer was 

 astatic. For damping non-aNtatic ones he had found it useful to 

 wind several turns of wire round the bobbin, and put them in 

 series with a few thermo-electric junctions warmed by the hand 

 and a key. In reply, Mr. Nalder sai I the galvanometer was 

 astatic, but the damping coil could be placed so as to act on one 

 pair of magnets more than on the other. A paper on a 

 portable instrument for measuring magnetic fields, with some 

 observations on the strength of the stray fields of dynamos, 

 by Mr. E. Edgar and Mr. H. Stansfield, was then read. The 

 instrument was described as an inversion of a d' Arsonval galvano- 

 meter, for the torque necessary to maintain a suspended coil 

 conveying a constant current parallel to the field gives a measure 

 of the strength of the field. The constant current is furnished 

 by a Hellensen's dry cell which the authors found remarkably 

 constant. The instrument consists of a coil of about fifty ohms, 

 wound on mica and suspended by two German silver strips 

 within a tube. A pointer is fixed to the mica, and a divided head, 

 to which the outer end of one strip is attached, serves to measure 

 the torsion. Within the head chamber is a commutator which 

 automatically reverses the current in the coil when the head is 

 turned in opposite directions from zero. Two readings may thus 

 be taken to eliminate gravity errors due to want of perfect balance 

 in the coil. Means are provided for adjusting and measuring 

 the tension of the suspensions. The constant of the instrument 

 was determined by placing the coil in the field of a 

 Helmholtz galvanometer, and found to be 0*293 P^^ i"- Any 

 other field is therefore given by 0*293 (« + i)d, where is the 

 angle of torsion in degrees, and n the multiple of 50 ohms in 

 series with the coil. Fields from 2 or 3 C.G. S. lines upwards 

 can be measured to about 2 percent, by the instrument, and even 

 the earth's field is appreciable. The authors have tested the 

 fields of dynamos at the Crystal Palace Exhibition and else- 

 where, and the results obtained are given in the paper. It is 

 noted that the stray fields of multipolar machines fall off much 

 more rapidly than those of two-pole dynamos as the distances 

 are increased, and that near edges and corners of the magnets 

 the fields are much stronger than near flat surfaces. The dis- 

 turbing effect of armature reactions on the strength of the stray 

 fields were measured, and the shapes of the fields observed in some 

 cases. Experiments on magnetized watches are described in the 

 paper. Mr. Whipplesaidihe Kew Committee wtre to someextent 

 responsible for the experiments described, for it was on their 

 account that the investigations were commenced. In connection 

 with the rating of so-called non-magnetic watches, it was neces- 

 sary to know what strength of fields they were likely to be sub- 

 jected to. The instrument devised for making the tests was a very 

 interesting one, and the results obtained by it of great value. 

 Mr. A. P. Trotter hoped the authors would supplement their 

 work by tracing out the directions of the fields of dynamos, and 

 he described a simple method of doing this by a test needle 

 used as an india-rubber stamp. The question of watches, he 

 thought, must be considered soon ; even non-magnetic watches 

 were stopped by being placed in strong fields, owing to Foucault 

 currents generated in the moving parts. Mr. Blakesley inquired 

 whether the instrument could be used in any position. He 

 thought three observations would be necessary to completely 

 determine any field. Mr. Stansfield, in reply, said they used a 

 pilot needle for showing the directions of the fields, and then 

 placed the coil accordingly. The instrument could be used in 

 any position, for the weight of the coil was only about 2 grammes, 

 and did not greatly alter the tension of the suspensions, which 

 was usually about 300 grammes. A watch with a brass balance 

 was not influenced by a field of 10 C.G.S. lines but seriously 

 affected by one of 40. — Mr. Joseph W. Lovibond read a paper on 

 a unit of measurement of light and colour. The paper was illus- 

 trated by many coloured charts, diagrams, and model^, and 

 several pieces of apparatus by which colour measurements could 



