ott me ‘ aida vr oe ate sp Oe ee eo ee ee ee ee ee ey A re eh eth ty ta tiled teas @ be od omy ay ~ FO ped bee idea yh Aimed eGo uyk ee LS ee he Beaty eR Real Oh) haat me th bed ‘ shat. sets ts betty bas etree 8 obebe dM ae - BWW tide the att oats [betta = Diet 4 M Prey ot way Pe ee er LE belle an . bot ea ie ‘ i a athe A Sy erent Fyne ery WN AW LMM hha bed aye -mebetbebraettraede Wade Pag 8 wrt re co bls tahe ia a oe ee or | ef ara eM ree Lr aration bar, BAL A 4! 1 Pr dek tae d en cee eo eee er’ Pat: tho tabs Row Poth wre peau tues Ce a yah a Pe Weer AA a TP ae kee ee Sarath whan rad hey dole dob beet , be iet © ALO De ee ee ee re er er sh Filed fo Yadeh 44.04.08 bewdee b ledeeo ded. 4 Nee Welt Gibre = eR oth do Ha ded are wa eran era be ‘ wie rere PCR RET Ler iC we Wr Unie be rar Pe Vpn na . Me 48 Meh: Ube ote dtd haAB thee cabs " taevae ee ee Pr AM et CN tn Bebe bbe 7 40) whee HAWN Cet Die Oda oh WO a gai bold Refined We dadalt late ' PO ke Diode that Cotbed a Mlb Dey, ne a CMe PLES ne We bP De wi “ t Da Fae tage inl Aura gata “ tele 48 8 Be ad ate a ten ote ae Woe Lede ade deted ad ha. bre bal ce rl ee ieee ee ony ore ne sr ee eres | Sater peal hae RE dot be hid det ded ed ae de et Cea Tot ee dig ts ty * “yh ‘aad Heo bed Hae aad a t+ eae ta alan Lhe ee bee beet oe abe det) bee ho ree a bate Lad dd ae didn “4 ASM te eas Meola dette DOMME dd Ah be ASO ID oi he bt ed » Wingo bahedemat mao “ id-® det oy 4 ae Ce ee eee oe etki a “ Nihal Had Gated mod uh theta ds Payoh t) Phar | PR er) Dedede ete etd dat Bh bie eee Oban sb thate we lee Ahh Chri orator ' 44 ‘ nt M4 oe wee | a) O80 hae tan rte Pia Hoe es the Lr ee a Hers 4 dus re ee ee ey eS tobi. DAL UE eee ee) de Aektn de vetit Daly 7») rie doth ON Hy 0M Bel dee el 4 et hy ae ie ee vibe (he dad eto ‘ , 6 ey Wo sa va Adee J i-8 ae a v5.4 i aa Sweet ». Pty ver tee ier Wn ice ny eye ie gree oe Preyer i ses it es Redon abe devs * wu VL deed thei don A tke wed " Mae ea Bas deb mite Poa misulepsitae iver ies 0d bul ga Lee res I) Hee WeR add) Rw MW hed deted Way ackea weenw bebe ican wict NeW 629 dednar ed ‘ iA Ode Ohaede *) 1 4 or baie lh oa if Doak deere eo, Lal A ores fae a thea whashoucsore 4 rh ; "4 gr 4 lidside Wao P' A dnt DE UE bed dele tet de Ld Vode ee ree ene, ee Re re) att wits. erp 1 riety ted ed dal Bee eRe UM oe rnc et vied weetek Medea HAs et br ew: ‘ sha ‘ ooh a awe Pichewdctvine’s alah 6 Nat ATW he te Pdr Ro, eh td be Mats Bod, euous Haile We thet de stamens 4 : 6d-eate Jobat: 8 8 toda td Mid ae Wr prerer wir ition Ouch Ved. mace she Gaeta v4 ; sated 40 ed ok ee dee tte WP Pod Ae dotted aey . dette. ate aast be Dvav0 ati bdenust as ‘ is ’ ‘4 i) AS aL Lt ee ere re Wea ae bends « aa he ve (ieee: we bd ode aba take ne ef llama 4 ‘ 9? nat on td A bait ied se ities ae ; Cn ee " in _ es Mt) sf Say ins i a Wa! me mit . ‘ é : ” He biWoltes- tka hes hold hat ei ra Wake eae i oo leh Wen ty Ww st ‘an Phe haghie aA Waa dey tts it sie Ain Deh On ies 4 i ee ie ta ie i mike 4 y bitstaeate ‘ q ] 4 J tiie i st et ‘ai ’ lee and " veh tayatia ‘te ici Rebath ort ee i PRM MEE ae BH ‘ea A a" Sse 2 Se Wim air he tae Pete & ae iw hi WH Hey) “4 WANT ean Y Per) 9 aah i Sit Bae 4 8a i? é a r : ES et eee a 9 % i. se rt + he tvies mt ori iisy: ry pate Bee rhe hit hey! aa ithe Win aie nai ati a DOP ath bof oo A aie Dem aWraas “4 ahaa, = teee re SA i) hh de et seas oy a inant . . wae op “et . ‘ i. ny 32 hea 4 & aa e 7 i) rot Ppl. bes ron eet e ip A a fish ft d: heck: Bathe: abr < CPA be Bee isk igh Mac mG *) eae te pon a le 4A iF ihe a able fray ea Saat tya a8 Had tas inset te lb Poabs | alt Won) de be ay in 4 vi rae Vath de had yy 2 ih + hy sige! he ‘i tin ai. Ry: uae et te Wail Make devas cn te Son Whey RA vbr dvas ie Bah Pa mn Wake rosa an she JAE ve Perdana * . baited y Say eT ee Cae en bs hs didn Hada ibe Aatis.t Caper eat Arak pe} wo oA Wali abe) 4) 4 idee 1c nite tad del Weer sh Pah dada val Ore. 9 (EAe ee Ca Lebar ie ata ort) we warstieire tae ee Bites te nee ’ “ 6 hee! Qa “wiv fat Aye ‘ Ve fecbonnta tare: tistan + hate! de yledt ee “itt os ie Tas vis Pade Ait VG dh Gye melt A. HW! Ae ede Me aie aad Leche he rte dad od Wad Lhe aiith jew fp Pet R hed Pe M$ BoA 0 a day Sth ela boil Hed» aes: He ria 4 ‘ ‘ wie aa mn edits 4 Satie suite ae ri a0" OP Aa Meade We Pale! Aad 2M ee Tt | ; p 4 Hy Abed Gldlic shied ait we WHE iz Fla has teitstras sn ? fy We eeede Bae de Ron itevead Ue ais Wea gs + tae \ ile Ry analog ae eee ¥: eta Ny Co janee 44 fy kaa ee fe sree Ot a sasjeby i a ele Py sere L phdeeat ths gad r \) Ave Wedelt 0405 0 baw rhea Aidala tral an tHe oh ar a ad Nedalt 1 ae ei deite j casi Cad ri idan on pad aera 4 a ni Ne arrerone| Oo ae nase ebenreyes abso Aree ony halt awd sf eet seed rhe Mae aw ile ¥ Be 4a ie oko | ‘ P oe ant 929.8 4): a5 Ken boas Paden Medubbia liar nade: ade +. , oes ty ey ate cel h a aa wea ec aD eine ae ‘ ee uae * ’ ide Ae 6d ete te “ idem J da bets teomaueg par he ‘ 1 i448 Wills 19 He toed ee Pied deg ca ha seh eas asi lind hee Oa ee ited q idgaa Lad Pee ee) WP Ww ded, ‘ «Eve Mn Pes N= wen odds “4 aa oa-W & ™ ‘ a a MES Lore cee 4) 4Ga8, ener : “ 24 bs awe € eG oe age . : ee ee i ylere hae ee is ‘ 5 Tr) i Md We ve ~~ ¢ a4 te eee 4 ‘ pew hatog bey a fe Lan we tered hed ye 8 0 ty deeb. beings Cred ‘icacaeaciea ts > ‘ soe iA ae Pieodor yer ee Ee ow . Cwee oe . OA rE he ead how gel da bad ad ‘ . . . aw * ’ “ 0 se Wao ole wid BS Pah Patera het bela iat 4eu ew d Uk Ge TAOW ree ok O WO on Wenn . a be a ted tallow dew "6 wet Wek Width wgued ah Fiat he bei ae no be id ee HU eT hells 7 rr | Woh Gow ded ey! bs | *) Hew 8 dodo adele Woviitto Wego owe. Roveewel dete tevak Be be eda) a Ae ed ed a Bed a8 Lene Leite ded Ee rite a FEM ei Aa =H, le sete We ee baw De Weds dot ee ey a a48 re ie ad oe he 48 weve itlen it ce eee a Patna sae re a ie Sei ees ha ie “tard dg a eéa ‘ ieet aod G id un | ewe PES PIC NST ae oa “4x inl Ga ah S185 baat: taRidae Wad i needa, pe eee. 2 ; ete hia GI ebadeds oy ce ee Ue CeCe totes ae rate el oa ye A a eo Cor et at Saeeee Nha te ov reinyideb Gaunt eh et eS a a) sda aie Mock } 4 head edo 4 ee Sh as ome ied be de Re Abd i we ie) dod ad vt BEM aed Oey i " on OKT Wilda Caan ay am ” Sevhok ar te Wade Stew ACR AO oe ed hag TE vgs eied A Wruht Wali don, detearienred satan gee 4 eres jad, wed ere ae —e Ve Ca) “oe Foe ea wey ~ = ws “ margins. 8 “i st is Surface smooth, hairs on margins. 5 ey! 3 7 ’ Surface smooth, hairs on margins or lobes. 4, F a 5 Surface smooth, hairs on lobes. 4A ee * if Quite smooth. Totals 36 34 2 Thus of the thirty-six plants originally bearing leaves with an intermediate surface only two remained unchanged. These thirty-six plants included the survivors (thirteen) of those descendants of A and B which originally produced leaves with an intermediate surface; the variations occurring in them are shown in Table V. 20 Miss E. R. Saunders. Table V. On a Discontinuous Detarled Analysis of the 13 Descendants of A and B, which originally bore Leaves with an Intermediate Surface. Number of plants. 13 Type of earlier leaves. 11 Surface intermediate. 3) 9) 39 39 3) 93 Table VI. a9 » daairs on lobes. Type of later leaves. Surface smooth, hairs on margins. hairs on lobes or margins. Detarled Analysis of the 45 Plants bearing Leaves with a Smooth Time of Number sowing. of plants. al 1 ha Aug. 4 I i 1 1 1 w (eral e re. J ; es 5) | 5 5 2 (i saal: cares March 4 a | 1 2 ea L Total . «se 45 Surface. Type of earlier leaves. Surface smooth, except for 1-2 hairs at the leaf-teeth ; hairs on margins. Do. do. do. Do. do. do. Do. do. do. Do. do. «do. Surface smooth, hairs on margins. a 55 hairs on lobes. ” 29 29 39 Surface smooth, except for 1-2 hairs at the leaf-teeth; hairs on margins. Surface smooth, hairs on margins. » 39 45 se hairs on lobes. Surface smooth, except for 1-2 hairs at the leaf-teeth ; hairs on margins. Surface smooth, hairs on margins. 9 39 2) ) bP) ” - hairs on lobes. Type of later leaves. Surface smooth, hairs on mar- gins (many). Surface smooth, hairs on mar- gins (few). Surface smooth, hairs on mar- gins or lobes. Surface smooth, hairs on lobes. Quite smooth. a ” Surface smooth, hairs on lobes. 39 : 9 39 +p) or quite smooth. Quite smooth. Surface smooth, hairs on mar- gins, Surface smooth, hairs on lobes or margins. Surface smooth, hairs on lobes. bP) ) 33 9) or quite smooth. Quite smooth. 2” 3) Surface smooth, hairs on mar- gins or lobes. Surface smooth, hairs on lobes. ”? by) LP) bP) or quite smooth. Quite smooth. Surface smooth, hairs on lobes. 3) 2 ” 39 ” or quite smooth. From the observations summarised in the foregoing tables, it will be seen that the cas¢s in which the whole number of leaves produced Variation occurring in Biscutella laevigata. 21 by one individual exhibit a fairly uniform degree of hairiness (or smoothness) are almost invariably those belonging to the extreme types; the plants are either very hairy (type I) or almost glabrous (types IIIc and the smoother forms of IIIb). All the plants in | which the leaf surface was originally free from hairs remained smooth, while out of the total number of hairy plants only two varied from the original type. It is those in which the first formed leaves are intermediate in character between these two extremes that the change from a more to a less hairy condition may generally be traced (types II, Illa, and the hairier forms of IIIb). Consequently we find that the continuous series of gradations from the condition of absolute smoothness to that of extreme hairiness, which may be observed upon examination of the leaves of a large number of seed- lings taken at random, is not met with in an equally large and hap- hazard collection of adult plants; among the latter certain forms have disappeared, and the types which obtain are more sharply marked off from one another. In fact adult plants fall into two groups; the type with leaves with an intermediate surface is not found, or occurs. as a rare exception. It follows, therefore, that a census compiled from a set of adult plants, and a similar record — obtained from the same individuals before the stationary point has been reached will not give concordant results ; in the former case the proportion of plants bearing leaves with a glabrous surface will be higher than in the latter. In order to ascertain the nature and amount of the variations occurring among the offspring of unlike parents, certain individuals which flowered in the summer of 1896 were intercrossed. The plants were placed under muslin covers in order to exclude insects, and the flowers were emasculated while still in bud before the anthers had dehisced, in order to prevent possible self-fertilisation, The seeds thus obtained were sown the same year after having been allowed to ripen for a few weeks; the character of the cross-bred seedlings is shown in Tables VII and VIII. : 9? Miss E. R. Saunders. On a Discontinuous Table VII. Classvficutton of 120 Cross-bred Seedlings. 7 Surfac Surface Surface “haley. interme. smooth. Totals. Number of seedlings derived from five 4 7 26 37 hairy plants x smooth (hairs want- ing or confined to the lobes) plants. Number of seedlings derived from five 5) 32 28 65 smooth (hairs wanting or confined to the lobes) plants x hairy plants. Number of seedlings derived from one 12 6 0 18 plant, surface smooth, marginal hairs numerous x hairy plant. & | Totals eeoee7eG oe eevee ee neo e088 S1 5A, 120 Table VIII. Classification of the same 120 Cross-bred Seedlings arranged in Families. Offspring Offspring of five Offspring of of one rather hairy plants fertilised five very smooth plants smooth by very smooth plants. fertilised by hairy plants. plant Plant Plant Plant Plant Plant Plant Plant Plant Plant Plant fertilised 1. Ze 3. 5. 1 3. 4, 5. by hairy Type of earlier leaves. Cc —A~A-—_——+ c— A— > plant. Surface hairy ...... — — 1 -— 8 2 — = — 8 12 Surfaceintermediate. 1 Shek 11 12— 6 8 6 6 Surface smooth, ex- cept for 1-2 hairs at the leaf teeth; hairs On Margins ...... 3 Boral 2— Pele = 3 4 2 — Surface smooth, hairs OY Margins .eciee Ze BM Po — ) 2B ee TO ae -~ Surface smooth, hairs on lobes ....000082 — Oro a i i i — Quite smooth....666. — = — i Eure OAS we apa LL tle A 3. 4 #18 Ll YW) i443 18 In all the “‘ very smooth” plants used as parents in these experi- ments, the leaf-surface was quite smooth, and if marginal hairs were present they were confined to the leaf-teeth. In the ‘“‘rather smooth” plant there were numerous hairs on the mar- gins and leaf-teeth. From the observations made upon this one generation of cross- breds, it would appear that when the extreme forms are intercrossed the offspring seldom exhibit the degree of hairiness characteristic of the more hairy parent; in most cases the first formed leaves corre- Variation occurring in Biscutella levigata. 23 spond to one of the glabrous or intermediate grades. In the one instance in which a plant not belonging to the extreme smooth type, but with numerous marginal hairs, was crossed with a hairy form, the general Jevel of hairiness in the offspring was very considerably higher. At the time of writing the character of the twenty-one hairy plants was unchanged. ‘The fifty-four plants bearing leaves with a smooth surface were still smooth, and the number of marginal hairs was gradually becoming less. The remaining forty-five plants originally bearing leaves with an intermediate surface were all tending to become less hairy, in fact the leaf surface was free from hairs in all those which had apparently reached the stationary point.* A few experiments were also made with the view of determining the character of the offspring in cases in which the parents resembled one another in texture. To this end certain individuals were placed under muslin covers, and either self-fertilised or crossed with others of the same type. Unfortunately none of the plants that were smooth set seed; four hairy individuals of the Genevese stock, however, fruited freely, and from them sixty plants were obtained, all of which showed the same degree of hairiness as the parents. It is of interest to compare these numbers with those obtained from the two hairy plants A and B (see Table II), which were freely exposed to insect visits; in the case of the latter only about 80 per cent. of the offspring belonged to the hairy type. Although the results tabulated in the preceding pages have been obtained from observations upon a comparatively small number of plants, they are, I think, sufficiently concordant to justify the follow- ing conclusion. The experiments went to show that a blending of parental characters as regards hairiness and smoothness occurs to a certain extent in the offspring of plants of dissimilar types, giving rise to intermediate forms. But this intermediate condition in respect of hairiness is only found exceptionally among full-grown individuals. For whereas in plants which at first are distinctly harry, the hairiness persists almost without exception, I have found that in nearly every case those plants which as young seedlings present an intermediate condition, assume, as they grow older, a more dis- tinctly glabrous habit. This change of character in the cross-bred seedlings which are originally intermediate, takes place gradually ; occasionally it does not occur until several leaves have been produced, but more often it is apparent as soon as the second and third leaves have developed. %* Though the seeds were all sown together, they germinated at such unequal intervals that the plants were at this time in very different stages of development. 24 Miss E. R. Saunders. On a Discontinuous HistoiogicaL Nore. A microscopical examination of the leaves revealed the presence of a histological feature of some interest. The general arrangement of the tissues follows the normal dorsiventral type ; it is the mesophyll, however, which claims especial attention, and exhibits a structural peculiarity which consists in the thickening and lignification of the cell walls in such a way as to form a latticed network. The meso- phyll cells of the cylindrical leaves of some species of Sansevierta exhibit a somewhat similar appearance as has been previously recorded by De Bary,* and figured by Henfrey.+ These bands of thickening give the characteristic lignin reaction with Schulze’s solution and with phloroglucin. In all respects, save for this modification of the walls, these cells resemble the rest of the mesophyll, and judging from the amount of chlorophyll contained in them, their capacity for amylogenesis is not less than that of the unaltered cells. The number and distribution of these cells varies considerably in different cases, but so far as I have been able to ascertain, their oceurrence is not correlated with any other structural feature. In order to determine their presence or absence in any given case, the leaves to be examined were allowed to rot in water until the epidermal layer could be easily peeled off with forceps from the under surface; they were then mounted whole, with the under surface uppermost. When sufficient transparency could only be attained by decolorisation, the leaves were placed in alcohol, and afterwards boiled for a few minutes in water or in dilute HCl, to facilitate the removal of the epidermis. The method of examination by sections was found to give unreliable results, since the distribution of the cells is so erratic that their non-occurrence ina number of sections affords no certain criterion cf their absence from the whole leaf. They may occur singly and remote from one another, or in groups in the meshes of the fibro-vascular network; or continuous layers of the spongy mesophyll or of the palisade cells may undergo this modification; in one case they may cover an un- interrupted area almost as large as the leaf surface, in another they may perhaps be confined to one side of the midrib, and be wholly wanting on the other, in fact, their distribution would appear to be entirely haphazard. In some cases they are present in the cotyle- dons, but more often they are absent from these organs, though they may be present in the later-formed leaves of the same plants; their presence in one leaf does not necessarily imply their occurrence in other leaves of the same individual. To take a single instance—nine leaves belonging to a very hairy plant were examined; in five of these the cells were abundant everywhere ; in two afew were present; and * ‘Comparative Anatomy of Phanerogams and Ferns,’ p. 118. + ‘Elementary Course of Botany,’ p. 483 (2nd edition). Variation occurring in Biscutella levigata. 25 in the remaining two they were absent altogether. Besides those of Biscutella levigata I examined the leaves of four other sub- species (?)—viz., B. raphanifolia, B. ambigua, B. lyrata, and B. awricu- lata, but I failed to discover any indication of tignification in the mesophyll of these plants. Addendum, July 28, 1897. Since the foregoing paper was communicated to the Society, I have myself visited the Val Formazza, and examined the character of the Biscutella plants in this locality, and also in one or two Swiss valleys. The results of these observations are given below :— Val Formazza, from the head of the valley down to the Tosa Falls (5500 feet). In this reach of the valley the plants are abundant everywhere— on the banks of shingle and sand and in the low-lying meadows near the stream, and on the grassy slopes of the surrounding heights, up to the Val Toggia on the one side, and as high as the lower limit of the Hohsand Glacier on the other. Both the hairy and the glabrous types were found, each variety often forming patches of varying size; such groups of dissimilar plants may occur side by side on exactly similar ground ; or, on the other hand, a small area may be occupied by both forms, which are indiscriminately mixed together. On the whole the smooth individuals were more numerous than the hairy, especially in the low-lying meadows near the river, on the steep slope up to the Val Toggia, and on the slopes on the opposite side of the river between the chalets of Morasco and Riale. In the above-mentioned meadows intermediate plants were also found, especially the smoother forms, and though very few in number compared with the extreme types, they were more numerous here than in any other locality which I had the opportunity of observing. Many of these intermediates were apparently young plants, and their comparative abundance in this spot may, I think, be explained by the fact that the season was a late one, and that consequently some individuals were ranked as intermediates, which had not yet reached the stationary point, and which would eventually conform to the smooth type. Val Formazza below the Falls and at Al Ponte (4200 feet). Here the plants were much fewer in number than in the upper reach of the valley ; immediately below the falls they were almost all smooth (no intermediates were seen); at Al Ponte the hairy and the glabrous types were found together on the shingle near the stream. 26 Prof. Ff. ©. Bower. Val Bedretto (All Acqua, 5265 feet). The plants were exceedingly abundant in this part of the valley, both on the lower grass slopes and close to the stream; in both places the great majority belonged to the hairy type. Intermediates of the more hairy kind occurred here and there, generally in patches. The very smooth type was not common. Val Canaria (Airolo, about 3900 feet). Here the plants, which were only moderately abundant on the grass slopes, were all hairy. Valley of the Rhone (at the foot of the glacier). A few plants were growing on the shingle 1 in the river bed, all very hairy. Valley of the Rhone (Ulrichen, 4380 feet). Only a very few plants were found, all very hairy. Valley of the Rhone (Eginen Thal). Plants numerous, both glabrous and hairy occurring together; a few of the hairier forms of intermediates were also found. Val d’Anniviers and neighbourhood of Berisal (Simplon). According to Mr. Bateson’s observations in the preceding year Biscutella plants were abundant in both these localities; in the former all the plants were very hairy, in the latter the hairy type predominated, but some hairy intermediates were also found. “Studies in the Morphology of Spore-producing Members. Part III. Marattiacee.” By F. O. BowER, Sc.D) F-i-8.. Regius Professor of Botany in the University of Glasgow. Received May 27,—-Read June 17, 1897. (Abstract. ) The memoir, of which this is an abstract, deals with the sori of all the four living genera of Marattiacesze; the development has been traced in Angzopteris and Marattia from the earliest stages to maturity, in Danea and Kaulfussia from such early condition as the material would permit. Some of the results from Danea have been already submitted to the Society in a preliminary statement.* One result of the investigation has been to demonstrate, as regards their development, the substantial unity of type of the sporangia in the four genera. In all of them a single “superficial parent cell ” of prismatic form is to be recognised embedded in the massive sporan- gium when young, not in a central position, but directed obliquely * ‘Roy. Soc. Proc.,’ vol. 59, p- 141. Studies in the Morphology of Spore-producing Members. 27 towards the centre of the sorus. By periclinal division this forms internally the archesporium, externally that part of the wall where dehiscence takes place. The tapetum arises, typically in them all, from the cells surrounding the archesporium. The dehiscence is in all by a slit in a radial plane, which may widen to a circular pore in Danea. In those sori where the sporangia are united laterally there is no annulus; it is present only where the sporangia are separate, as in Angiopteris. An interesting feature is disclosed by estimates of the potential spore-production of the single average sporangium in the four genera; the results in round numbers are, in Angtopteris 1,450, in Danea 1,750, in Maraitia 2,500, in Kaulfussta 7,850. It is to be remembered that the usual numbers in Leptosporangiate ferns are 48—64.; in some Leptosporangiate ferns (Osmunda) the number may rise to 500. I have ascertained in Gleichenia, however, that the number may be as high as in Angiopteris. This large potential out- put of spores goes parallel with the broad base of the sporangia; in fact, the Eusporangiate condition is that best adapted for maturing large numbers of spores in the individual loculus. Frequent deviations from the type have, however, been observed, as well as variations of size and mode of segmentation of the sporangia, and it is not possible in certain cases to refer the whole sporogenous tissue of one sporangium to a single parent cell. A special study of the irregularities has been made in Danea, in which genus they are most marked; incomplete septa are frequent, and the sporangia are of very unequal size. The main features have already been noted in the preliminary statement on that genus, where it has been pointed out that comparison of the details with those of the septate anthers of some Angiosperms shows that there is a remark- able resemblance between the two cases. Similar irregularities have been noted, though less commonly, in Kaulfussia, and Marattiva, and rarely in Angiopteris. Those fossil Marattiaceze which are best known as to the details of the sorus have been compared, and the substantial similarity of the sori in certain cases to those of the modern genera recognised. The facts from fossils and from the modern Marattiacee have been made the basis for a fresh discussion of the theoretical question, whether the synangium is or is not a result of coalescence of sporangia? It is concluded that the paleophytological evidence leaves the question open as to the priority of existence of forms with synangia, or with separate sporangia, in the Marattiacee. Notwithstanding that writers of authority have treated the question as decided, that the synangia are a result of fusion of distinct sporangia, it is held with some persistence that it is still open; the paleophytological evidence is inconclusive, while the comparative evidence from the living 28 Mr. C. 8. Tomes. On the Development of genera will not only accord with, but appears actually to support a view of septation. For the analogy with septate anthers, where septation must have occurred, and the similarity between the details of these and those in Danea, and especially the partial septations in both, make it appear probable that in this genus progressive septation has taken place. It is thought probable that progressive septation has been a feature, at least where the sori are elongated, as in Danea. But the question is left over for future discussion whether or not a similar septation, © rather than coalescence, may be accountable also for the origin in the first instance of a circular sorus with a plurality of sporangia united together as in Asterotheca, or in Pecopteris unita. “Qn the Development of Marsupial and other Tubular Enamels, with Notes upon the Development of Enamel in General.” By CHARLES S. Tomes, M.A., F.R.S. Received July 12, 1897. (Abstract. ) It was pointed out by my father, the late Sir John Tomes, that the enamel of marsupials was peculiar in that in the whole class, with the solitary exception of the Wombat, the enamel is freely pene- trated by tubes which enter it from the dentine, and are continuous with the dentinal tubes at the junction of the two tissues. This character is met with sporadically in other mammals—for example, in the Jerboa among rodents, in the Shrew among insectivora, and notably in the Hyrax, in which animal the free penetration makes its enamel look quite like that of a marsupial. _ Whilst there is a large literature upon the development of ordinary enamel, little or nothing has been written about that of tubular enamels. : The outermost portion of marsupial enamel is always devoid of tubes, and the extent to which the tube system exists varies greatly in different members of the group, so that the same enamel organ is obviously capable of forming either tubular enamel or enamel with solid prisms. Moreover, the sporadic reappearance of tubular enamels amongst mammals who have for the most part lost this character, and its occasional occurrence in a rudimentary condition as an abnormality in man, point to its not originating in any manner fundamentally different from that of ordinary enamel development; and it is claimed that the study of its development in marsupials affords the clue to the real nature of enamel development in all animals. Marsupial and other Tubular Enamels, &. 29 The nature of the question renders it impossible to convey in brief space the grounds upon which the conclusions have been arrived at, but they are— 4. That the special cells of the enamel organ Gongcests) do not themselves calcify. That they each furnish from their free ends outgrowths or pro- cesses which are continuous with their own plasm, and which may be traced through the entire thickness of young enamel. That one ameloblast furnishes the whole length of an enamel prism. That the fibrillar outgrowths, previously more or less correctly described by other observers in other enamels, but apparently not appreciated at their full importance, do calcify from without inwards in such a manner that an axial canal is left uncalcified. Hence the canals of marsupials are in the centres of the prisms, and not, as supposed by Von Ebner, in the interspaces of the prisms. And that towards the completion of the full thickness of the enamel the central axis is no longer left soft, but the whole calcifies into a solid prism. . It is claimed that other enamels, for instance human enamel, ealcify in the same way. It has long been known that short pro- cesses hang out from the ends of the ameloblasts, and these, having first been described by my father, are generally styled Tomes’ pro- cesses ; and also that the earliest formed layer of enamel is per- - forated, so that acids will peel up a perforated membrane from its surface during its development. Longer fibrils have also been detected by Andrews, Williams, and others; but so small a thickness is occupied by these structures, and the full solidification of the prism follows so close upon the heels of any change in the direction of calcification, that the true nature of these structures has not been detected. But in marsupial enamel, owing to the tubular condition which is so very transient in human and other mammalian enamels being per- -manently retained, the problem is presented under conditions more favourable for elucidation. Hence it is my belief that all enamels alike are formed by the cen- tripetal calcification of fibres furnished by the ameloblasts, and that tubular enamels are nothing more than the perpetuation of a stage which is passed through, though only for a brief period, by every solid enamel prism. This view serves to explain the occurrence of the various forms of tubular enamel which are found in fish, in some of whom— e.g., Sargus—the reverse order of things is met with—that is to say, the prisms first formed near to the dentine are solidly calcified, but as their growth goes on the later-formed portions become tubular, so that in the completed enamel there appears to be a system of tubes entering it from its free surface. 30 Prof. R. Boyce and Dr. W. A. Herdman. In certain cartilaginous fish there is a combination of both of these arrangements of tubes, from the dentine and from the sur- face, and sundry other apparently anomalous conditions are met with. But if the views advocated in this paper be accepted, all diffi- culty in accouuting for these arrangements, very difficult to explain from any teleological standpoint, disappear, for they become merely slight variations or arrests at different stages of a process common to all enamels during their formation. “On a Green Leucocytosis in Oysters associated with the presence of Copper in the Lencocytes.” By RUBERT Boyok, M.B., Professor of Pathology in University College, Liverpool, and W. A. HerRpMaAN, D.Nc., F.R.S., Professor of Zoology in University College, Liverpool. Received July 9, 169%. In the course of an investigation upon oysters under normal and abnormal conditions, upon which we have been engaged for the last two years, and upon which we propose to submit to the Society a detailed memoir during next session, we have come upon a pheno- menon which we regard of such considerable importance that we desire to publish a briet record of our observations and experiments, as we believe they may prove of interest to other biologists who are engaged in work on the micro-chemistry of the cell. The phe- nomenon we have now to describe is the presence of large quantities of copper in certain green leucocytes found in a diseased condition of the American oyster. The oysters suffering from this leucocytosis are always more or less green, but’ must not he confounded with ordinary green gilled oysters, where the colour is due to a totally distinct cause. History. Green oysters have been known from an early period, and there are various historic cases on record* of people having been poisoned by eating green oysters, and of the oyster merchants being put upon trial because of the deleterious nature of their goods: Periodically green oysters have been suspected or convicted of being coloured with copper, and just as often it has been proved by competent authorities that copper has nothing whatever to do with the green colour. This difference of opinion in the past has undoubtedly been * An interesting historical survey of the subject up to 1866, was given by the late Mr. Arthur O’Shaughnessy, in the ‘Annals and Mag. Nat. Hist.,’ ser. 3, vol. 18. On a Green Leucocytosis in Oysters. dl largely due to the fact that the observers worked with different kinds of oysters. Some have investigated the celebrated “ Huitres de Marennes” (a form of Ostrea edulis), and have found that while having dark blue-green gills, they were still in a perfectly healthy state, that they contained very little copper, and that some iron was present in the pigment. All that is perfectly correct, but it does not enable us to draw any conclusions in regard to other green oysters. There are evidently several kinds of greenness in oysters, and whereas some may be due to normal and healthy processes, others must be regarded as abnormal or diseased conditions. It is the latter, in our experience, that contain the copper. As early as 1835, Bizio showed that certain oysters he obtained at Venice contained copper, and he attributed (1845) their bluish- green colour, and that of the Marennes oyster, to the presence of that metal. Two subsequent discoveries have thrown a certain amount of probably undeserved discredit upon Bizio’s work. These are (1) the determination by Fredericq and others that a certain small amount of copper is present normally in the hemocyanin of the blood of crustaceans and molluses; and (2) the excellent work of Lankester* and others on the Marennes oysters which established the normal, healthy condition of the greenness, and the absence in that form of any copper beyond the trace due to hemo- eyanin. We now think it very probable, in the lhght of our recent experience, that Bizio was dealing, in the case of his Venetian oysters, with the same copper-bearing green pigment that we have met with. About 1880 Ryder investigated some green oysters in America, and from his description of what he found we cannot doubt that he had before him the same kind of green American oyster (Ostrea virginica) that we have been examining. He showed that the green colouring matter was taken up by the amceboid blood cells, and that these wandering cells containing the pigment were to be found in the heart, in some of the Pend vessels, and in aggregations in “eysts”’ under the surface epithelium of the body. He describes the colour (in the ventricle) as a “delicate pea-green,” and states that it is not chlorophyll nor diatomine: he suggests that it may be phycocyanin or some allied substance.+ So far as we are aware there has been no work¢ since Ryder’s, * See Professor Lankester’s memoir on “ Green Oysters,” in the ‘ Quart. Jour. Micro. Sci.’ for 1886, which gives an excellent discussion of ene subject so far as the Marennes oyster is concerned. + Ryder’s papers are in the ‘ U.S. Fish. Commission Reports and Bulletins’ rom 1882 to 1885. { Except Carazzi’s passing allusions to our work in ‘ Mitth. Zool. Stat. Neapel’ for 1896. His own investigations were made upon other kinds of oysters. 32 Prof. R. Boyce and Dr. W. A. Herdman. dealing with what we described a couple of years ago as the green leucocytosis in the American oyster. Many papers, to which we do not refer, have appeared dealing with other kinds of green oysters, but they do not affect our present subject. In January, 1896,* we referred briefly to what appeared to be an inflammatory condition, accompanied by a pale chalky-green colour, which we found in some American oysters relaid at Fleetwood, on the Lancashire coast; and at the Liverpool meeting of the British Association, last September, in discussing various kinds of greenness in oysters, we referred to this diseased condition, in the following terms :— ‘There is, however, a pale greenness (quite different in appear- ance from the blue-green of the ‘‘ Huitres de Marennes”) which we have met with in some American oysters laid down in this country, and which we regard as a disease. It is characterised by a leuco- cytosis, in which enormous numbers of leucocytes come out on the surface of the body, and especially on the mantle. The green patches visible to the eye correspond to accumulations of the leuco- cytes, which in mass have a green tint. These cells are granular and amoeboid. The granules do not give any definite reaction with the aniline stains, and; so far, we have not made cut their precise nature.” Finally, towards the end of last year, in the ‘Supplement to the Twenty-fourth Annual Report of the Local Government Board,’ Dr. Bulstrode corroborated our statement as to the presence of the pale green disease from the examination of specimens from Truro and Falmouth. Dr. Thorpe stated in the same Report that he had found that some green oysters from Falmouth,+ sent to him for examina- tion by Dr. Bulstrode and Mr. R. Vallentin, contained notable amounts of copper, in some cases as much as 0'02 grain per oyster, while the amount normally present is only 0°006 grain. Dr. Charles Kohn has kindly, during the last year or so, made a number of analyses for us of different kinds of oysters——“‘ natives,” ‘¢ Marennes,” Dutch, and American—and whereas in most of these he has found the copper to be present only in small quantities, on the average agreeing well with the amount (0:006 grain) usually stated as present in the tissues of the normal healthy oyster, in some green Americans which we gave him recently for the purpose he has found a very much larger amount of copper. These circumstances induced us to reopen, in our investigations, the question of copper in certain green oysters, with the results that are detailed below. * Report for 1895 on the Lancashire Sea-Visheries Laboratory (‘ Trans. Biol. Soc.,’ Liverpool, vol. 10, p. 158). + Obtained from a creek, which is locally supposed to bring down copper, and the water of which was found on analysis to contain some copper. On a Green Leucocytosis in Oysters. 33 The Green Leucocytosis. We first noticed this diseased condition in the autumn of 1895, in some ordinary American oysters (“blue points”), belonging to the species Ostrea virginica, which had been imported into Liverpool and relaid near Fleetwood, in the estuary of the Wyre. Since then many hundreds (probably several thousands) of American oysters have been examined by us, and we have seen all degrees of the leuco- eytosis. It manifests itself in patches and streaks of green on the mantle and other parts of the integument, in engorgements of the blood vessels, especially of those that ramify over the surface of the viscera, and in masses of green-coloured leucocytes in the heart. This green condition, although much less frequently seen in “ natives ” (0. edulis), is occasionally met with there also, and we have recently had some specimens from Falmouth with very well-marked green hearts, due to an accumulation of leucocytes laden with green granules in the ventricle. Such hearts are of frequent occurrence in the diseased American oysters; after death the mass of leucocytes subsides to the bottom of the cavity, leaving the clear plasma above. It is thus easy to demonstrate that the colour is due to the leuco- cytes, and to the leucocytes alone. The blood of these oysters contains a great variety of more or less colourless and more or less green and granular corpuscles, all of which may be termed leucocytes. They are apparently all amceboid wandering cells, comparable to the colourless corpuscles of the blood of higher animals. The larger and (probably) older of the leuco- cytes are very coarsely granular and very opaquely green. It is these that give the colour in bulk. We find them in masses in the heart, in both auricle and ventricle, in the vessels, where they are | sometimes so abundant as to engorge or inject certain parts of the system, in the lacunar spaces of the connective tissue of the mantle and other organs, and also in the more solid parts of the tissues wandering amongst the other cells, wedged into the epithelium and coming out in great numbers on the surface of the body. Some of these iatter, when found in the ectoderm and on the surface, are ~ very markedly eosinophilous ; those in the vessels are not so markedly so. When stained with osmic acid the granules of the leucocytes become black. After treatment with fat solvents, however, some of the leucocytes are still very granular. In sections which have not been stained, the granules of the leucocytes have a distinctly brown colour, recalling the appearance of the granules in the liver cells in unstained sections in cases of pernicious anemia. We opened many hatches of eae oysters, 100 at a time, and in all cases where the green tint was present in the mantle, heart, or VOL. UXII. D 34 Prof. R. Boyce and Dr. W. A. Herdman, vessels we found the accumulations of leucocytes. From 120 oysters we chose the six greenest and the six whitest. Dr. Kohn analysed these two sets of six for us, and found that the green contained between three and four (3°7) times as much copper as the white. This shows that it is not merely a redisposition in the body of the copper, due possibly to the hemocyanin, but that there is an absolute increase in the amount present in the body. We also found that the greenest parts of the body contained far more copper than corresponding tissues which had no green deposit in them. Not only then do these green oysters contain a largely increased amount of copper, but we have also shown that the copper coincides in its distribution with the green leucocytes, and, conse- quently, we regard the copper as the cause of the green colour. We then passed on to a more minute examination of the pigment and to _ histo-chemical reactions. Chemical Reactions, The Green Colouring Matter—The greenest portions of the green oysters were snipped out and dried on the water bath. The dried pewdered residue was treated with alcohol, ether, chloroform, benzene, turpentine, xylol, but these reagents failed to extract the colouring matter; we concluded, therefore, that the pigment was not of the nature of a lipochrome. On the other hand, the pigment was readily soluble in dilute acids and in alkalis; the addition of am- monia gave rise to a distinct bluish tint, and fresh pieces of the green oysters reacted instantly with ammonia, with the formation of a beautiful blue. . We next determined whether the pigment was due to iron or copper. The dried residue treated with dilute hydrochloric acid and potassic ferrocyavide gave a marked red reaction, thus indi- cating the presence of copper, and it was then found that very small quantities of the green colouring matter treated with dilute hydrochloric acid were sufficient to produce a well-marked deposit of metallic copper upon polished iron. In several instances a deposition of copper occurred when a piece of polished iron was laid upon a green patch on the surface of the mantle of a fresh oyster, dilute hydrochloric acid having been previously used to moisten the mantle. Control experiments were made with the whitest portions of the American oysters and with natives, and traces only of copper were found. These results have been also quantitatively controlled by Dr. Kohn. A series of histo-chemical reactions were then carried out. For the purpose the oysters were hardened in absolute alcohol, and pieces were then imbedded in paraffin, great care being taken that On a Green Leucocytosis in Oysters. 385 ‘every reagent was perfectly pure, firstly, with regard to the absence -of copper or iron, and, secondly, that no acid was present; thus, for example, commercial turpentine may give a distinctly acid reaction, and this would be sufficient to remove the copper. If sections -were imbedded in gum—and often the best results were obtained by this method—the tissues were allowed to remain for as short a time as possible in distilled water and then transferred to perfectly ‘fresh neutralised solution of gum-arabic, and allowed to remain in it for only a short period, The pigments appeared partially soluble in water. Comparatively thick sections were cut, in which the distribu- tion of the green colour could be seen with the naked eye. These were placed in absolute alcohol in every case before proceeding to test, The reagents which we employed were potassic ferrocyanide, ‘15 per cent. solution,* freshly prepared ammonium-hydrogen sulphide, and pure hematoxylin. Potassic Ferrocyanide.—Sections were taken from absolute alcohol and passed into distilled water for a moment in order to remove the alcohol. They were then placed in the potassic ferrocyanide solu- tion, when the portions previously green assumed a red colour; this reaction set in immediately. The presence of a 0°5 per cent. solu- tion of hydrochloric acid added in equal quantity to the ferrocyanide -solution previous to use (as recommended by Macallum for iron) tended to hasten the reaction, and in some cases was necessary in order to obtain it. The sections were then washed in distilled water, dehydrated in absolute alcohol, cleared in cedar oil, and mounted in Canada balsam. The red coloration was found located to the masses of leucocytes, and the individual leucocytes themselves were of a faint yellowish-red colour. In the cases of the very granular pigmented ‘leucocytes the granules assumed a distinct red-brown colour. In this way the distribution of the leucocytes and of the vessels which contained them was mapped ont. Very beautiful preparations of the engorged green vessels were obtained by partially dissecting the mantle in the fresh oyster so as to expose the ramifying vessels, then -hardening in alcohol, and subsequently treating with ferrocyanide solution, when the vessels assumed a well-marked red colour; beautiful ‘results were also obtained by ammonia. Fresh blood obtained from the heart in which vast numbers of the green leucocytes were present also gave a red reaction with acidulated ferrocyanide solu- tion. Control bloods from white oysters gave an exceedingly faint or no reaction. * For the sake of uniformity we finally adopted the strength of solutions given by Macallum in his paper on the “ Distribution of Assimilated Iron Compounds,” “Quarterly Journal of Microscopical Science,’ 1896. D2 36 Prof. R. Boyce and Dr. W. A. Herdman. Ammonium-hydrogen Sulphide—Sections taken out of the alcohol and placed in this solution instantly gave a marked dark yellow- brown reaction wherever there were green patches. This reagent is more striking in its results than the potassic ferrocyanide, and very good cover-slip preparations of the blood can be obtained, the cor- puscles staining dark yellow-brown. Heematoxylin—We were led to use this reagent from knowledge of its reaction in the case of Weigert’s nerve-staining method. The results are most striking. Sections placed in a watch-glass of distilled water, to which a few crystals of pure hematoxylin are then added, begin at once to assume a distinct blue colour in the place of the previous green; this occurs whilst the solution itself remains free from colour, and therefore whilst the quantity of hematoxylin dissolved must be very minute. Microscopic examination shows the corpuscles dark blue, and the vascular network beautifully differenti- ated. The connective tissue and gland cells and nuclei remain un- stained, or occasionally show a very faint blue reaction, most marked immediately around the vessels. This reaction appears to us to be as specific for copper as Macallum showed it to be in the case of inorganic iron. Just as in the test-tube, so in the cell, a blue-black reaction is obtained not only with iron (as in the state seen in the liver cells in pernicious anemia) but also with copper. It therefore follows that hematoxylin is a most sensitive test for either metal, and that consequently in the outset it is necessary to determine whether copper or iron is present exclusively in the cells, and to which of these elements the reaction is due. Tron is found in the ash of the oyster, and the green coloration of the Marennes oysters has been attributed to it by Carazzi and others. In the case of the green oysters which we have examined, Dr. Kohn found, in addition to the copper, traces of iron—the iron was, how- ever, far below the copper in quantity. In the detailed and valuable paper of Macallum, previously referred to, a series of histo-chemical reactions are described in order to demonstrate the presence of iron in cells, and he with others distinguishes two forms, organic and inorganic. The latter, like, we presume, the iron in the liver cells in pernicious anzmia,* gives an immediate reaction with potassic ferro- cyanide and dilute hydrochloric acid, and as Macallum has shown,} a dark blue with pure hematoxylin. But the organic iron. behaves differently, giving, according to Macallum, a yellow colour with hematoxylin, and requiring previous treatment with dilute nitric, sulphuric, or hydrochloric acids in alcohol before a Prussian blue reaction is obtained with acidulated potassic ferrocyanide, or prolonged * We have obtained an immediate blue reaction with hematoxylin in the liver in five cases of pernicious anemia. + ‘Report British Association,’ Liverpool, 1896, p. 973. On a Green Leucocytosis in Oysters. 37 treatment with ammonium sulphide before any dark coloration is ob- tained with that reagent. Nowit will be observed that all our reactions were immediute, taking place directly on the addition of ammonium sulphide, or potassic ferrocyanide alone, or aided by a trace of acid. The copper was therefore present in a condition analogous to the inorganic iron, or at least so loosely combined with the cell proto- plasm as to be readily discharged, but in none of these cases did we get any indication of inorganic iron, except in the case of the contents of the alimentary tract of the oyster. When the sections were treated with 3 per cent. nitric acid in alcohol for half an hour the green colour disappeared, and then neither the copper reaction nor the striking reactions with ammonium sulphide and hematoxylin took place. Subsequent treatment of these sections with acidulated potassiec ferrocyanide, and again washing in dilute nitric or hydro- ehloric acid, yielded a general and very faint Prussian blne reaction, in which the nuclei of the gland cells were more markedly blue than the leucocytes. If the method is reliable it shows that traces of iron are present in the cells in addition to the copper, but it is the organic iron. Some oysters gave this Prussian blue reaction more markedly than others; this was the case with some F'almouth “natives.” Hemocyanin containing copper has been shown to be an important constituent of the blood in many of the invertebrata, taking the place of hemoglobin. We have examined the blood of very many oysters, and only in two instances, and these in green oysters, have we thought that the plasma became very faintly blue on exposure to oxygen, whilst, as previously indicated, qualitative tests either failed to give any indication of copper or, at most, only a very faint reaction, and even in these cases the reaction appeared confined to the leucocytes which were present in the plasma. The ash, how- ever, of the white oyster yields about 0°006 grain of copper, and it is probable that minute traces are present in the plasma as hemo- cyanin. The cause, then, of the presence of the copper in such abundance in the green leucocytes is very obscure. The quantity of copper in the green leucocytes themselves varies, as our histo- chemical reactions demonstrated; some corpuscles could be found which were conspicuous by their red reactions on the addition of potassic ferrocyanide, whilst others in the same preparations only gave very faint indications, and occasionally a cel] could be seen which gave a marked Prussian blue reaction instead of the red. Conclusions. Onur results demonstrated the presence of copper in comparatively large quantity in the green leucocytes, chiefly in the American oyster, but also in the “ natives ’’ from Falmouth and other localities. 38 Mr. J. W. Swan. Stress and other Liffects We have shown that the colour was in proportion to the amount’ of copper present, and that the colourless leucocytes contained only traces of that metal. The deposition of the copper in this large quantity appears to us to represent a degenerative reaction; it was accompanied by a most striking increase of leucocytes, which tended to distend the vessels and to collect in clumps, phenomena which are abnormal in our experience in the oyster. The presence of the copper in the leucocytes in these cases might. be compared to that of the iron which is met with in some of the leucocytes in cases of old hemorrhages, pernicious anemia, or in other cases where iron is set free. We are not prepared to state whether copper in the food can bring about the condition, but cer- tainly we have abundant evidence to show that it can occur where no copper mines or other evident sources of copper are present. We are inclined to suggest that the increase of copper may be due to a disturbed metabolism, whereby the normal copper of the hemocyanin, which is probably passing through the body in minute amounts, ceases to be removed, and so becomes stored up in certain cells. Our results also show that hematoxylin is a most valuable reagent, not only as Macallum has shown in the case of iron, but also in that of copper, and that care must be taken to distinguish between the two reactions; and this must be especially the case in those invertebrata where copper plays an important réle in the physiology of the blood. “Stress aud other Effects produced in Resin and in a Viscid Compound of Resin and Oil by Electrification.” By J. W. SWAN, F.R.S. Received May 17,—Read June 17, 1897. (PLates 1—4.) While making an experiment with the object of finding the degree of resistance to puncture offered by paper coated with a soft com- pound of resin and oil, when placed between the secondary terminals of an induction coil, the tension being. regulated by a spark-gap in a parallel branch of the circuit, observed that on the passage of a spark at the spark-gap, while no spark passed between the paper- separated terminals, a sudden roughening or puckering of the previously smooth surface of the coating took place. A number of experiments were made with the object of ascertain- ing the nature of the action which produced this effect, and these led to further experiments and to results which, though closely related to well-known phenomena, possess features of novelty and interest. produced in Resin, §e., by Electrification. 39 It was found that clear Bordeaux resin in a viscid state (viscidity being brought about either by heat or by the addition of resin oil) is responsive to the mechanical stress consequent on electrification by non-luminous discharges; and if it is so acted upon while in the solid state, and afterwards superficially softened by heat, there results a new kind of electric discharge figure, analogous to the dust figures of Lichtenberg and Lord Armstrong, but showing some remark- able peculiarities which throw additional light on the mechanism of air- conveyed electrical discharges, and on the location and nature of the stresses imparted to the dielectric. Jascertained that a smooth sur- face of resin is retentive of an electric charge to an extraordinary degree, that after more than two months the lines of an electric discharge figure, as developed by heat, and as further developed by the acci- dental attraction of atmospheric dust to the electrified parts of the surface, were still attractive of dust in a discriminative manner, no change being observable upon re-dusting either in the arrange- ment or definition of the lines of electrification as originally developed. The apparatus employed consisted of an induction coil or a Wims- hurst machine, and a supporting stand, the rod of which carried two clips and a stage, the supporting part of the stage being made of | strips of thick plate-glass, and the rest of wood, The clips held conducting wires, which passed through bent glass tubes, and went to the secondary terminals of the induction coil, or to the conductors of the Wimshurst machine; the discharging arms in either case con- stituted an adjustable spark-gap in paralle] with the wires ending above and below the stage. The stage terminals were balls, discs, or points; the pairs employed in different experiments varied in size and form, and the pair used together were sometimes dissimilar. The resin was the colour of amber; in some of the experiments it was used ina solid state, but fused to the form required for experiment ; in other experiments it was softened tosemi-liquidity by the addition of 20 per cent. or more of resin oil, the mixture being made by fusion together of the resin and oil. The compound with 20 per cent. of oil has the consistency of treacle at a temperature of 20° C.; at 12° C, it is nearly solid, yet plastic enough to yield to the mechanical stress-action generated by the projection upon its surface of an electric discharge of the kind employed in the experiments. At the higher temperature the viscid liquid is well suited for showing the great disturbance produced by repeated discharges, and when at the lower temperature it is convenient for observing the more persistent forms of the figures produced on the surface by single discharges under various conditions. When it was required that the stress figures should be permanent, resin either alone or with not more than 2 or 3 per cent. of resin oil was used. 40 Mr. J. W. Swan. Stress and other Effects The dielectric was either contained in glass basins or spread as a coating of 0°5 to 1 mm. thick upon glass or mica plates, and in a few cases.on copper plates. Also plates consisting wholly of resin were in some instances used. The effect of a spark passing at the spark-gap, when one of the stage terminals is suspended over, and at a certain distance from the viscid resin and oil mixture contained in a basin, the other being in contact with a metallic disc under it, is to produce an evanescent figure on the surface. The character of the figure depends on :— 1. Whether the terminal over the dielectric surface is positive or negative. 2. The form and size of the + and — terminals. 3. The distance of the upper terminal from the surface of the dielectric. 4, The potential] and character of the spark at the spark-gap. Typical Efects——The most regular and characteristic stress figures are obtained when the spark-gap is adjusted so as to prevent the passage of a spark or visible brush through or over the dielectric, but allow a non-luminous discharge to take place of only slightly less strength than would be necessary to produce a brush discharge visible in the dark. A typical effect is obtained when the spark-gap is 25 mm., and the positive branch from it terminates in a brass ball of 8 mm. diameter hanging centrally over, and 4 mm. from the surface of, the dielectric (80 per cent. resin and 20 per cent. resin oil at 20° C.) contained in a glass basin 150 mm. diameter and 15 mm. deep, the negative wire being brought to a dise of metal 100 mm. diameter under the basin, or to a disc of tinfoil attached to the underside. On breaking the primary circuit by means of a mercury break with a trigger action (the spark-gap having been momentarily short-circuited while the primary circuit was closed), and the con- sequent passing of a single spark at the spark-gap—no visible dis- charge occurring between the ball and the dielectric—there suddenly breaks out on the surface of the viscous liquid a star-shaped figure formed of deeply furrowed, closely clustered, outward-branching rays, extending from a circular frill near the centre to the margin of the liquid. The figure gradually dies down, and on the surface becoming smooth it can, with slight variations, be reproduced again and again by repeated breaks of the primary circuit. . If the commutator is reversed (the spark-gap being momentarily closed while the primary circuit is re-made), then on breaking the primary circuit as before, a figure characteristic of the negative con- vective discharge is produced. This figure 1s much smaller and weaker than the positive one; most frequently it consists of a produced in Resin, §c., by Electrification. 4] circular, or nearly circular, band or ring, more or less indented in outline, enclosing leaf-like rays which tend towards the centre. These are relatively broader and less branching than the rays of the positive figure, and they are characterised by vine their outlines in relief, while the rays of the positive figure are sunk below the plane of the surface. When the sissteiaedtian is strong, the ring enclosing the rays stands up as a frill in considerable relief. Effect of the Form of the Terminal.—The character, both of the positive and negative figures, is greatly affected by the form of the discharging electrodes. When the upper terminal is a metallic disc of 25 mm. diameter, hanging in a plane parallel to, and 2 mm. above, the dielectric surface, and the other terminal is a 50-mm. metallic disc supporting the basin, complicated, but nearly symmetri- cal, figures of great beauty are obtained. A metal point opposed to a metal point, or a metal point above and a small metal ball below, give smaller figures of more elementary forms, having the general characteristics of the larger figures. Balls and points as terminals tend to produce circular eu in which the rays converge to, or diverge from, one centre. This rule applies with fewer exceptions to the + figure. The negative figure, even when produced by a discharge from a brass ball, is frequently a combination of sectors, whose centres are not far apart, and are concen- tric with the centre of the group. The effect of this is to produce a figure of nearly circular outline broken by more or less regular inden- tations. If the ordinary vibrating contact-breaker be employed instead of the trigger-break used in the foregoing experiments, the rest of the arrangement remaining as described, larger and more complicated effects are obtained. When the ball above the liquid is positive, the resin and oil being at a temperature of 20° C., on breaking the primary circuit the first effect of the make-and-break is the produc- tion of the characteristic star with arborescent rays; the repetitions of the impact which instantly follow indent the lines of the figure more and more deeply, and result in the effacement of the more regular figure, and the development of a large and turbulent movement of the liquid, tending to its division into two masses: a central mound with a flat or concave top and a concentric ring. At first the two masses are joined by radial ridges, but these gradually thin, and (if the discharges at the spark-gap are continued) eventually break down and leave the central mound and the embracing ring com- pletely separated. During the progress of the action the ring por- tion is driven outwards, and when the limit of outward movement is reached, there is a subsidence of the more violent agitation, the oater ring becomes somewhat smoother and flows inwards; this is followed by a recurrence of the repellent action, and a repetition of. the last 42. Mr. J. W. Swan. Stress and other Effects phase of the phenomena described. Figs. 1 and 2 represent in profile the appearance at the middle and fina! stage of the action. Fie. 1.—Positive. Nearly Maximum Effect. ERG pee ee Fia. 2.—Positive. Maximum Effect. _ Ifa metal ring (of 90mm. diameter) is substituted for the disc, similar but more sharply defined effects are obtained. When the arrangement described is varied by making the ball over the basin negative, instead of positive, on breaking the primary circuit there is less displacement of the viscid material consequent on repeated interruptions of the circuit. To obtain a characteristic effect of repeated negative dis- charges, the spark-gap should be widened to 50 mm., and to prevent sparking over the edge of the basin, it should be at least 150 mm. diameter. There is then formed, immediately under the ball, a concavity the counterpart in size and curvature of the ball, and outside this there is the general figure, somewhat faint and tremulous, which seems to be a complication of the positive and negative figures, the characteristic positive figure encircling the more distinctly nega- tive portion. Fig. 3 is a profile view of the negative displacement. The extent of the effects described is considerably modified by the degree of viscosity of the liquid, and this can be controlled by tem- perature. Fie. 3.— Negative. Maximum Effect. = Corresponding but modified phenomena are produced by means of the Wimshurst induction machine. When, for example, an 8 mm. metal ball connected to the + arm of the discharger hangs 15 mm. from the surface of the resin and oil mixture of the consistency of treacle contained in a large basin, and the.— arm of the discharger is con- nected to a metal disc of 100 mm. diameter under the basin, the dis-. charger balls being 75 mm. apart, the result of continuously working the machine is the production of a turbulent motion, attended by the produced in Resin, §¢., by Electrification. 43 formation of vortices into which the upper stratum of the liquid pours downward, carrying adherent air with it, while an equal and opposite movement takes place from below, producing worm-like eruptions of spirally twisting or wriggling jets of liquid at the sur- face. If the margin of the dish beyond the surface of the liquid is thinly smeared with the viscid liquid, the film breaks up into dew-like beads. With a metal ball of 150 mm. diameter, 6 mm. above the surface of the liquid and a proportionately larger disc basin and depth of liquid, acted upon by a machine of moderate power, the spark-gap being adjusted so as to prevent sparks passing from the ball to the liquid, and the tension such as to keep up a strength of electrification of the surface only slightly less than that which would cause disruptive dis- charges to pass, a column of liquid rises and connects the overhanging ball and the surface of the liquid in a manner strongly suggestive of water-spout phenomena. The ascent of the column of liquid is followed by the descent of numerous thin surrounding streams, and these keep up a regular ada of upward and downward circulation. Fization of Characteristic Ejffects—In order to fix the various forms produced in the viscous mixture of oil and resin, an experiment was made with nearly pure resin, rendered plastic by heat, and cooled to solidity while under the action of electric discharges. It was found to be difficult to carry this out in a satisfactory manner, but it suggested a reversal of the procedure, viz., the electrification of a surface of resin in a solid state, and the subsequent development of the stress effect by rendering the surface superficially plastic by heat. This mode of operation resulted in the production of permanent relief and intaglio figures, corresponding to Lichtenberg’s dust figures, to Brown’s photographs,* and to the dust figures and photographs more recently described by Lord Armstrong. The preparation of the resin surface for the production of the permanent figures requires care. The method I employed is as follows :—A thin glass basin was filled with a mixture of resin and © per cent. resin oil, the oil being added to lessen the tendency of the resin to fracture on sudden change of temperature. The resin, melted in a metal pan, was poured through a filter of muslin into the basin, while embedded in small shot and raised to the fusion temperature of resin. On slowly cooling, the basin being meanwhile covered with a plate of glass or an inverted basin, the resin solidifies with a smooth bright surface. The resin-filled basin was partly covered, on the bottom, by a disc of tinfoil, and was placed centrally * “Phil. Mag.,’ vol. 26, p. 502. . t ‘ Electric Movements in Air and Water, with Theoretical Inferences,’ by Lord Armstrong, C.B., F.R.S. London: Smith, Elder and Co., 1897. a Mr. J. W. Swan. Stress and other Effects on a metal disc in contact with one of the wires from the spark-gap of the induction coil, the other wire, ending in a point, disc, or ball, overhanging the surface at the distance of a few millimetres as in the experiments with the viscous material described. On the production of a single spark at the spark-gap by means of the trigger-action mercury contact-breaker, a charge is conveyed to the resin. The peculiar distribution of this charge, and that it is attended by strong and enduring mechanical stress, can be made manifest either immedi- ately, or many hours afterwards, by slightly warming the surface of the resin. The result is a deeply impressed figure, having the same character as the figure produced on the viscid dielectric. These solid figures, if carefully developed, show much fine detail; unfortu- nately, this is not capable of complete illustration by photographs. Difference of depth in the grooves is not fully indicated, neither is there represented adequately a peculiar burring of the margin of the grooves, especially in the negative figures, their edges rising slightly above the plane surface, as though the resin had been finely carved. Persistency of the Stresses—The persistency and fixity of the electrification of the resin surface, determining the form and character of the eventual figure, are very remarkable. If the development of the effect of the charge is delayed for twenty-four hours there is but little difference in the result from that which follows immediate development. Dust Figures and Stress Figures combined.—It was found that the heat-developed figures attracted dust from the atmosphere, and thus formed, accidentally, a combination of a stress figure and a dust figure. The attracted dust gave clear indications of electrification beyond the limit of the stress figure, and brought out features of detail which helped to explain the nature of the electrification. Analysis of the character of the figures in this respect is still further helped by combining a modification of the dust process of Lichtenberg with the stress effect described in this paper. The modification referred to consists in allowing the dust—the mixture of red lead and sulphur proposed by Lichtenberg—to be drawn up to the elec- trified surface from a cloud of dust. This assists in the selective appropriation of the two substances, giving a redder colour on the negatively electrified portions of the figure, and a yellower colour on the positive portions; and showing what the stress figure alone does not show with equal clearness, how inseparable are the + and — actions. The best effect is obtained by applying the dust process before development by heat. Figs. 4 and 5 (Plate 1) represent characteristic forms obtained by positive and negative discharges with a metal ball electrode above the surface and a metal dise below, fig. 4 being + and fig. 5 —. Figs 6 and 7 are corresponding figures obtained with discs above and below; fig. 6 being + and fig. 7 —. Roy. Soc. Proc., vol. 62, pl. 1. — Swan. Pie. 4. Roy. Soc. Proc., vol. 62, pl. 2. Swan. Roy. Soc. Proc., vol. 62, pl. 3 Swan. produced in Resin, §¢., by Electrification. 45 - Figs. 8 and 9 are additional examples of negative discharge figures with larger disc electrodes. An excellent dust figure, in which the result of + electrification is strongly developed, is obtained by suspending, face downwards, an electrified resin surface in a very thin fume-cloud produced by burn- ing magnesium ribbon. The fume should be enclosed in a box, or under a glass shade, and an hour should elapse for the coarser parti- cles to subside before the introduction of the electrified surface. Sulphur in a state of sublimation can also be used in the same way with good effect, especially for very small areas of electrification, where microscopically fine development is required. On the whole, however, I have found nothing better than red lead and sulphur ground separately to very fine powder, and used very dry in a dust- ing box, the electrified surface always being downward when exposed to it. With the object of finding the degree and kind of interaction between the positive and negative electrification produced on opposite sides of a solid dielectric, interposed in the path of a single discharge, the following experiment was made:—A thin plate of glass was coated on both sides by dipping in melted resin, this was electrified by bringing the secondary terminals of the induction coil, arranged as in the experiments already described, to opposite sides of the plate. The terminals were brass balls 8 mm. diameter, placed in a vertical line, the + above, the — below the plate in a horizontal position ; the + ball was 1 mm. distant from the upper surface of the plate, and the — ball was in contact with the under surface. Under these conditions when an 8-mm. spark passed at the spark-gap, reciprocal figures of a very interesting character were produced, a + figure on the upper surface and a — figure on the under surface. To enable photographs to be taken of these figures without in- terference, the experiment was repeated with the variation that a plate of ruby glass coated with resin on both sides was used instead of clear glass. The latent figure was first developed by means of the red lead and sulphur cloud, and afterwards the stress effect was brought out by heat. Fig.10 shows the form of the figure on the + side, and fig. 11 that on the — side. When these double figures are viewed by transmitted light, it is seen that the interior — rays on one side, coincide with the inner ends of the outward streaming + rays on the opposite side. That the depth of penetration of the charge which produces these figures is very small is shown by the almost complete discharge effected by washing the electrified surface with water. The experiments seem to show that when an electric discharge takes place through air, its propagation is attended by a structural arrangement of the air brought under the influence of the discharge, 46 ‘Dr. C. I. Forsyth Major. and that when a dielectric like resin is interposed in its path, some of the characteristics of the form into which the electrified air has been thrown are transferred to the resin surface as an electric charge, generating the stresses and other inductive effects which result in the dust and stress figures. Experiments corresponding to those described made in an atmo- sphere of carbonic acid gas at normal atmospheric pressure, and in air at pressures lower than the normal, show that the character of the figure imprinted on a dielectric in receiving an electric charge through a gaseous medium is largely dependent on the density of the atmosphere conveying the charge; greater density tending to con- centration of the figure and attenuation to diffuseness. With an air pressure supporting 85 mm. of mercury, the other conditions being such as would have given at normal pressure a characteristic + star figure, there was diffuse electrification of the resin surface, but there were no rays. | «On the Brains of two Sub-Fossil Malagasy Lemuroids.” By C. I. ForsyrH Mayor. Communicated by Henry Woop- WARD, LL.D., F.R.S., V.P.G.S. Received April 6,—Read June 3, 1897. (PLATE 5.) The casts here described and figured have been moulded from the brain-cavities of the skulls of two sub-fossil Lemuroids from Mada- gascar, the descriptions of which I have already published. For com- parison with the brains of living Lemuroids the figures published by P. Gervais* are the best adapted for the present purpose, since they, too, are drawn from moulds of the brain cavity, and give on one plate a good general idea of the variations of the Lemur brain, 1. Globstlemur Flacourti, Maj. The larger of the two casts was taken from the skull briefly de- scribed by me at the meeting of the Zoological Society of London, June 20, 1893.+ In its general contours, as viewed from above (fig. 1), the brain of this form, for which I now propose the name of Globilemur Flacourtt (g. n. et sp. n.), approaches most to that of the smallest members of the family (Lemuridee), viz., Microcebus,+ both being remarkably broad * Paul Gervais, “‘ Mémoire sur les formes cérébrales propres a l’ordre des Lémurs,. accompagné de remarques sur la classification de ces animaux,” ‘ Journal de Zoologie,’ vol. 1, 1872, pp. 5—27, Pl. 2. + ‘Zool, Soc. Proc.,’ 1893, pp. 582—835, I P. Gervais, loc, cit., fig. 7, Pl. 2. Poy. Soc. Proc.Vol. 62.Plate 5. Forsyth Major. “Fig. 6. West, Nevaman lith. On the Brains of two Sub-Fossil Malagasy Lemuroids. 47 in their posterior moiety and suddenly attenuated anteriorly. Apart from the Sylvian fissure, the brain surface of Microcebus is perfectly smooth, whilst the cast of the fossil shows a greater complication than in any other known Lemurid. Thisis in accordance with what might have been anticipated, Globilemur being larger than any living Lemurid, and, as Broca states : ‘‘ Un cerveau qui grandit doit se plisser sous petne de déchoir*”’; this, in my opinion in fact, means that for economy of space plication is resorted to as a means of increasing the surface. | In the arrangement of its convolutions (fig. 2), the fossil departs likewise from what is known of Lemurid brains, and approaches rather more to what is presented by some of the larger Cebide and Cercopithecide. In Lemurids the fissures and the corresponding con- -volutions show a tendency towards a longitudinal arrangement, quite different from the more radiating direction exhibited by the fossil, Its Sylvian fissure (s.f.), on the other hand, corresponds in its more vertical direction to what we find in Lemurids, and in this respect departs more from the Old and New World monkeys, though less from the former than from the latter. The character mentioned is in relation with the development of the occipital lobe, the Sylvian fissure being always more horizontally directed in those brains in which the occipital lobe is well developed and in which, as a consequence, the cerebellum is covered. In fact, in Globilemur, the cerebellum is much less overlapped than in the monkeys. In the lesser development of the frontal lobes we find a further agreement with Lemurids as compared with monkeys, and equally so in the more macrosmatic character of the brain of Globilemur, as revealed by its voluminous olfactory lobes. I shall not enter into farther particulars as it is never safe to attempt to make out the exact homoiogies of the fissures in a cast of the brain cavity. Moreover, in this case, I find that the two sides of the hemispheres do not agree in every respect, owing partly to the incomplete condition of the skull and partly to the difficulties encoun- tered by the artist in the moulding. II. Megaladapis madagascariensis, Maj. The second cast, from the brain-cavity of Megaladapis madagas- cariensis, is in many respects the very opposite of Globilemur. First, -as to size,—from the dimensions of the respective skulls, the size of the first named animal (Megaladapis) may be approximately calcu- lated as double that of the last (Globilemur), whilst in bulk the brain * Paul Broca, “ Anatomie comparée des Circonvolutions Cérébrales. Le grand lobe limbique et la scissure limbique dans la série des Mammiféres,”’ ‘ Revue d’ Anthropologie,’ IT, vol. 1, 1878, p. 413, a5 Dr. C. L. Forsyth Major. of the smaller animal exceeded that of Megaladapis. So far as can be judged from the cast (fig. 4), the hemispheres were much less convoluted than in the large existing recent Lemurids, their fore- and hind-parts being apparently almost smooth in the fossil form. The reduced proportions of, or, more properly speaking, the absence of, the occipital lobe, is testified by the cerebellum remaining uncovered. . But the most remarkable character is exhibited by the anterior beak-like continuation of the hemispheres (fig. 5, b), which presents in section a triang ular form with a broad, flattened base and a trenchant superior margin. The corresponding part of the skull has been elsewhere described,* when it was shown that the correspond- ing constriction of the brain cavity is due to the enormously developed frontal sinuses protruding into the anterior portion of the cerebral and olfactory fossz. The optic nerves will help us by indicating the orientation in Hie curiously shaped brain (fig. 3 and fig. 6). From a comparison of the inferior part of the brain of Megaladapis with that of an Indris (fig. 7), it will be seen that in the former the frontal lobes are absent, and the part of the hemispheres situated in front of the optic nerves is represented by scarcely anything but the posterior part of the before-mentioned beak (fig. 6, 6), which continues aateriorly to form the olfactory tract which is equally reduced. Hven in croco- diles (fig. 8), the fore part of the hemispheres, anterior to the optic nerves, appears less reduced than in this Mammal. Little information can be obtained as to the anterior portion of the tractus and the bulbi, as it was not possible to mould this portion of the narrow channel running between the internal walls of the frontal sinuses; nor could this unique skull be bisected. As far as can be made out, the canal in question widens in proximity to the cribriform plate, so as to form the chambers for the lodgment of the olfactory bulbs. * “On Megaladapis madagascariensis, an extinct gigantic Lemuroid from Mada- gascar, &c.,” ‘Phil. Trans.,’ B, vol. 185, 1894, pp. 25, 26. + A somewhat similar conformation obtains in the Whales, to which my attention has been drawn by Sir William Flower, who described it in Balena mysticetus, where “the two somewhat dilated chambers for the olfactory bulbs are divided from the cerebral cavity by a canal which runs for a distance of 83 inches and is 13 inches wide and from 3 tolinch high.” ‘On the Greenland Right-Whale (Balena mysticetus), by D. F. Eschricht and I. Reinhardt; Appendix by the editor, William Henry Flower, London, Ray Society, 1866; W.H. Flower, ‘An Introduc- tion to the Osteology of the Mammalia,’ 3rd ed., London, 1885, p. 220. See also Ant. Desmoulius, ‘Dict. Class. d’Hist. Nat.,’ vol. 6, 1824, p. 372, s. v. “Event ” (Balena australis); F, Cuvier and Laurillard, in ‘Cuvier, Legons d’Anat. Comp.,’ seconde édit., vol. 2, 1837, p. 303 (Balenoptera); Otto Késtlin, ‘Der Bau des knéchernen Hoptess in den vier Klassen der Wirbelthiere,’ Stuttgart, 1844, pp. 16, 17, 18, 89 (Balena australis—Balenoptena borealis). On the Brains of two Sub-Fossil Malagasy Lemuroids. 49 As has been pointed out in the description of the skull of Megala- dapis,* its post-orbital region is remarkably elongate in the lateral parts, in a manner quite unusual amongst Lemuroidea, and for parallels of which we have to look amongst Carnivora, and especially Insectivora (e.g., Centetes). An external and superficial examination of the skull might lead to the belief that this elongation has resulted in an anterior elongation of the brain-cavity as well. But as we have just seen, in Megaladapis the elongation in question is brought about by the development of air sinuses, whilst the cranial cavity is on the contrary shortened, as well as narrowed. Although in this skull the satures are almost entirely obliterated, it is obvious that in the elongation of the lateral parts of the post- orbital region, the orbits and the alisphenoids participate as well as the frontals. This is well shown by the fact that, whilst in Lemurids generally, as well as in monkeys, the passage for the optic nerves from the internal cavity to the orbits, of which we speak as the optic foramen, is a very short one—very oblique in the former, almost parallel in the latter,—we find in Megaladapis that the second pair of nerves traverse a canal of no less than 243 mm. length, before appearing at the outer side of the skull, in the orbits. So that, in lien of a foramen opticum, we have here a camnalis opticus. The united foramina rotundum and lacerum anterius form likewise a canal of about 21°5 mm. length. When describing the skull of Megaladapis, I endeavoured to show that its peculiar low condition is not primitive, but pseudo-primtive (Fiirbringer), that is to say, that it has been brought about by a “ retrogressive evolution,” or a retrograde metamorphosis, if the last term be preferred. If any further proof were needed for this asser- tion, it would be furnished by the conformation of the brain, as described above, for I trust that no anatomist will maintain that this was the primitive condition in Lemuroids. It may fairly be predicted that, when we come to know the skulls of very young specimens of Megaladapis, they will show a much closer approach to the ordinary Lemurid type in the conformation of the brain cavity and its walls, and the gap between the young and the adult in this respect will prove to be wider than perhaps in any other known Mammal. How- ever, in the Insectivora and most of all in Centetes, we find also a very great difference between young and adult in the relative size and conformation of the brain (the brain being even absolutely smaller in the old), whilst the least divergence is to be found in Marsupials on the one side, in Man on the other, and this obviously for opposite reasons. Apart from what has been pointed out about the analogy of Megal- * Lac. cit., pi 16: VOL. LXIT. a 5) aie Drs. L. Mond, W. Ramsay, and J. Shields. adapis with the Whales, in the elongation of the anterior part of the brain-cavity, corresponding to the tractus, no instance of a similar reptilian-like conformation of the brain is known to me amongst Mammalia, if I except the Amblypoda, especially the Dinoceratide, the brain of which ‘‘ was proportionally smaller than in any other known Mammal, recent or fossil, and even less than in some reptiles. It was, indeed, the most reptilian brain in any known Mammal... . The cerebral hemispheres did not extend at all over the cerebellum or the olfactory lobes.”* EXPLANATION OF PLATE. Figs. 1—3.—Figures of cast of brain-cavity of Globilemur Flacourti, Major, Pleistocene, near Nossi-Vé, 8S.W. Madagascar. Original specimen pre- served in the British Museum (Natural History). 2/8rds natural size. Fie. 1.—View of brain, seen from above. 5, 2.—Side view of same (s.f., Sylvian fissure). », 3.—View of same, seen from beneath (0.2., optic nerve). All drawn 2/3rds natural size. Fies. 4—6.—Figures of cast of the brain-cavity of Megaladapis madagascariensis, Major (2/3rds natural size) ; Pleistocene, Amboulisatra, S.W. Mada- gascar. Fie. 4.—View of brain, seen from above (b, beak-like projection in front). 3) 0.—Side view of same. », 6.—View of same, seen from beneath (0.7., optic nerve). All drawn 2/8rds natural size. Fie. 7.—Brain of Indris (seen from beneath), recent (0.n., optic nerve) ; capied from Grandidier. », 8.—Brain of Alligator (seen from beneath), recent (0.z., optic nerve). “Qn the Occlusion of Oxygen and Hydrogen by Platinum Black. Part IL” By Lupwie Monp, Ph.D. F-R‘S., WILLIAM Ramsay, Ph.D., F.R.S., and JoHNn Surenps, D.Sc., Ph.D. Received July 21, 1897. (Abstract.) The heat of occlusion of hydrogen in platinum black was deter- mined by saturating the platinum black with hydrogen, extracting as much of this as possible at 184° C. by means of the pump, and then readmitting it again whilst the experimental tube was placed in an ice calorimeter. By proceeding in this way, errors due to the pre- existence of oxygen in the platinum black were avoided, and it was * O. C. Marsh, “Dinocerata. A Monograph of an extinct Order of gigantic Mammals,” ‘ Monographs of the United States Geological Survey,’ vol. 10, Wash- ington, 1886, pp. 538, 54. Occlusion of Oxygen and Hydrogen by Platinum Black. 51 found that 68°8 K (6880 g-calories) were evolved per gram of hydrogen occluded. It is shown that the arguments put forward by Berthelot in favour of the existence of the compounds Pt;)H, and Pt;,H; are not justified. According to Favre there is a difference between the behaviour of palladium and platinum to hydrogen, inas- much as when hydrogen is admitted fractionally, in small portions at a time, the heat evolved in the former case is constant, whilst in the latter it becomes less and less. This difference is apparent only and not real, and is due to the presence of oxygen in the platinum black. In order to determine the heat of occlusion of oxygen in platinum black, a great many experiments were made to try to remove the oxygen, which is always present, without destroying the occlusive property of the platinum, and so obtain platinum black which would per se occlude oxygen directly at the temperature of the calorimeter, and thus eliminate all corrections for the simultaneous occurrence of other reactions. Several reducing agents were emploved, including sulphur dioxide, carbon monoxide, ammonia, methyl alcohol, and formic acid in the state of vapour and in dilute solution, and it was found that, although the oxygen was removed, the reducing sub- stance or its products of decomposition were occluded by the plati- num black, and were just as difficult to remove as the oxygen itself; and, further, the volume of gas given off, derived from the reducing agent or its decomposition products, was approximately equal to the volume of oxygen originally contained in the platinum black. In most cases this was about 100 volumes. An extended series of experiments is described showing how platinum oxygen and platinum hydrogen can exist in the presence of each other. If the quantity of hydrogen which is theoretically necessary to remove all the oxygen in the form of water be admitted to platinum black, then, instead of removing all the oxygen first with formation of water, the hydrogen only removes the oxygen from the platinum black with which it first comes into contact, and imme- diately takes its place. The heat of occlusion of oxygen in platinum black was finally measured both directly and indirectly in the following ways. Plati- num black fully charged with hydrogen was exhausted at 184° C. to remove as much of this gas as possible. The experimental tube was then placed in the calorimeter, and oxygen was added in small quan- tities at atime. From the experiments on the co-existence of platinum oxygen with platinum hydrogen, the heat evolved during this process was known to be partially due to the formation of water and partially to the occlusion of oxygen. The vacuum in the apparatus remained perfect up to a certain point, when the presence of a slight excess of oxygen caused the pressure to increase. On now admitting oxygen up to full atmospheric pressure, a further small quantity of oxygen was BE 2 52 © Mr. J. Norman Lockyer. On the Appearance of the occluded, and the heat evolved represented the true heat of occlusion of this quantity of oxygen. Indirectly, the same value was obtained by charging the platinum black up fully and alternately with hydrogen and oxygen, and finally with oxygen. The amount of oxygen really occluded in the last charge, and independent of that which had gone to form water, was then found by exhausting im vacuo at a red heat. The difference between this quantity and the total amount of oxygen used is a measure of the oxygen which formed water with twice as much hydrogen by volume. Knowing these quantities, the total heat evolved, the heat of formation of water, and the heat absorbed on the removal of hydrogen, we have all the data for Spears the heat of occlusion of oxygen. : In a similar way the amount of heat absorbed per gram of oxygen removed was calculated from the data obtained during the penultimate charge. The mean value for the heat of occlusion of oxygen, from the direct and indirect measuremeuts, which did not differ much from each other, is +110 K (1100 g-calories) per gram. This value referred to 16 grams of oxygen is +176 K, which is almost identical with Thomsen’s measurement of the heat of formation of platinous hydrate Pt(OH)., viz., +179 K. This agreement suggests the possibility that the two phenomena may in reality be identical, the necessary water being always present in platinum black dried in vacuo. The paper concludes with some speculations on the nature of the occlusion of gases. “ On the Appearance of the Cleveite and other New Gas Lines in the Hottest Stars.” By J. Norman Lockyer, C.B., F.R.S. Received June 15.—Read June 17, 1897, Introductory. In my recent paper on ‘“‘ The Chemistry of the Hottest Stars,” * I left for future discussion the spectra of those stars apparently at or near the apex of the temperature curve, for the reason that in them the lines of known gases do not show very great variations, while the enhanced lines cease to be of service as a criterion of temperature. 1 pointed out, however, that there were several lines, as yet of unknown origin, which are strong in some of these stars and weaker in others, and that the study of these might eventually help us in classifying such stars and arranging them in temperature * © Roy. Soc. Pioc.,’ vol. 61, p. 185. Cleveite and other new Gas Lines in the Hottest Stars. 53 order, but that before attempting to use the unknown lines in these inquiries it was important in the first instance to discriminate, if possible, between gaseous and metallic lines. Until this point was ‘investigated the relative behaviour of the lines of hydrogen and cleveite gases near the upper temperature limit could not be satis- factorily discussed. The work has now been carried on a stage further, and in the present paper I propose to give the results of the inquiry into (1) the appearances of the lines of gases, both old and new, in the spectra in question, and (2) the most probable sequence of tempera- ture in the stars under discussion. The Spectral Lines by which the Sequence of the Hottest Stars can be determined. In the former paper I stated the sequence of certain stars, both of | increasing and decreasing temperature, as determined chiefly by the enhanced lines of iron and the lines of the cleveite gases. At the junction of the two series I provisionally grouped together Bellatrix, ¢ Orionis, y Urse Majoris, \ Tauri, and y Pegasi, pointing out that their spectra were not quite identical and might afterwards be separated when the criteria had been further studied.* Further inquiry has shown that y Pegasi may be regarded as practically identical with Bellatrix, while 7 Urs Majoris and \ Tauri differ from it chiefly in the general haziness of the lines; no attempt has been made, therefore, to separate these stars from Bellatrix. Other stars included in the present discussion were 6 and e Orionis. At the top of the ascending series of stars I placed Rigel and ¢ Tauri,t and, among others, at the top of the descending series were 8 Persei and a Andromede. The sequence of the still hotter stars can, therefore, be deter- mined by an investigation of the varying intensities in their spectra of lines which appear; also in stars on one side or other of the temperature curve. The principal lines utilised in this inquiry are as follows :— * €Roy. Soc. Proc.,’ vol. 61, p. 180. + This is one of the most extraordinary spectra which has been met with in the Kensington series of photographs, as I have already pointed out (‘ Roy. Soc. Proc.,’ vol. 61, p. 184). While the lines of hydrogen are fairly sharp and not very broad, many of the lines, especially those of the cleveite gases, are broadened almost into invisibility. On the meteoritic hypothesis this is explained by the great differences of velocity and direction of the meteoritic streams, the special broadening of the lines of the cleveite gases indicating that these gases are chiefly concerned in dis- turbances at high temperatures. On account of the indistinctness of many of its lines, Tauri is omitted from the present discussion. Nor the Appearance of the On man Lockyer. J Mr. o4 + 2 4 = ul Rnd be 3 7 Cleveite and other new Gas Lines in the Hottest Stars. 55 3933°8 Ca(K) 4173°2 unknown. 39648 Gas X 4179°0 unknown. 4009'4 ,, | 4340°6 He. 4026°3 He 4267°6 unknown. 4088°7 unknown | 4388'1 Gas X. 4128°6 Si 4131:4 Si The accompanying map shows the sequence of spectra in the hottest stars deduced from the behaviour of the above lines in passing from the stars of increasing temperature to stars of decreasing tem- perature, and includes also some of the typical stars on both sides of the curve, namely, a Cygni, 7 Leonis, and Rigel, and # Persei, a Andromede, y Lyre, and « Canum Venaticorum, in the order pre- viously determined. In each case the intensities of the various lines are indicated by their thicknesses, so that the variations in passing - from star to star are plainly shown. The wave-lengths of the lines and the origins of the known lines are shown at the bottom of the map. The map enables us to discuss the relative behaviour of each of the lines, and to notice which thin out or become more intense as the temperature changes. It will be seen that when one set of lines becomes very faint or disappears, another makes its appearance or becomes intensified. The map thus shows the most probable sequence of spectra among the stars near the acme of temperature as deduced from the changes of intensity of the lines given above. The Variations of the Cleveite Gas Lines. Comparison of the Principal Lines of Helium and Gas X.—In dis- cussing the appearance of gas X in relation to helium, it is necessary to deal with the subordinate series in each case, as the only line of the principal series of helium (\ 3888°785) which falls in the photo- | graphic region considered coincides with a hydrogen line, and cannot therefore be compared with the line of the principal series of gas X, which does come within range. Taking the lines 4471°6 and 4026°3 as representing helium, and 4388:1 and 4009°4 as representing gas X, the comparison shows that :— 1. Gas X does not vary absolutely with helium. _ 2. Gas X increases its intensity at a different rate from that of helium. : 3. When helium is at about a maximum so is gas X. The maxi- mum of gas X is, however, very short lived, while that of helium extends very considerably. 56 Mr. J. Norman Lockyer. On the Appearance of the These differences are shown on the map, and they fully accord with the laboratory work, which indicates that helium and x are to be regarded as distinct substances. Comparison of the Lines of the Subordinate Series.—In the above investigation it has been found that, in tracing the progress of gas K through the stars of increasing and decreasing temperature in the photographic region, the relative intensities of the lines of the different series are changed from those tabulated in the laboratory. The lines of the principal series, as indicated by the line at 39649, are no longer the strongest, but become of secondary importance as regards intensity, whilst the first subordinate series now takes the pre-eminent position, and the second subordinate series nearly dis- appears altogether, being only represented very feebly near the point of highest temperature. In the following tables, drawn up by Mr. Shackleton, will be found a statement of the relative intensities of the principal, first sub- ordinate, and second subordinate series of gas X and helium. Relative Intensities in Stars of increasing Temperature of the Lines in the principal and subordinate Series of Gas X. Pane ‘ Ist subordinate | 2nd subordinate Star rincipal series series series : (A 3964°9). (A 4388'1). (4437°7). TEU) 0 cae ee 5 ri a | Vie crew nc sive iacea 2 2 4 3 —- AP AVE OINIS ia ee eles oe 2 P trace os i OA (13 0S eRe at 2 1 — AU SANT Yease iets: sia) /o etm, oe — — — Ma OTiOMisees sis os —_ — | Relative Intensities in Stars of increasing Temperature of the Lines in the lst and 2nd subordinate Series of Helium. 1st subordinate | 2nd subordinate Star. series series (A 4471°6). (A 4121°0). is + tee ee | He UU GRT RS oc ote atel ds 10 ESSE Laranctst se laiet pis als ole 6 2-8 A AIBOUIS Ss cs oce serene’ os 2 BUCY STU ice, sates aise wipye iL y Cygni ee arereceveve —- MO TIONISY Sas acceso — Cleveite and other new Gas Lines in the Hottest Stars. 57 Relative Intensities in Stars of decreasing Temperature of the Lines in the principal and subordinate Series of Gas X. de : 1st subordinate | 2nd subordinate Principal series Star. , series series (A 3964'9). (\ 4388°1). (4437-7). 1 oe ere an Oy Daye c.- eos cceees Castor eeeveser7 0e ee ee PrOCYOD .. 2.2... 000 Arcturus eseeevev ee e808 [1] | wren 1 tL eco fet a esl = DREW Geos es ne oe 08 Relative Intensities in Stars of decreasing Temperature of the Lines in the lst and 2nd subordinate Series of Helium. | 1st subordinate | 2nd subordinate Star. series series (A 4471°6). (4121°0). REVIGUTIX «co's ce ca ck 00 10 4, RSPEESOY 2 fe ee es 3 2—1 PARI Dia ai aise is ow oe uf iL Sirius .. 4 a — a PEED Gils 6c 0 « ba se we - — — HTOCVORN ss s'Jies cess os PUVOUUTUS | in ois ale 06-000 The above detailed investigations show that while helium and gas X behave differently as regards their appearance in stars, the constituent series of each, so far as we can at present study their behaviour, do not exhibit any remarkable differences. Thus in the stars of increasing temperature there is a steady increase of intensity of the three series of lines of gas X, and of the two series of helium lines, while in the case of cooling stars there is a decrease in the intensity of each series. In the case of gas X it will be seen that the principal series is not intensified to the same extent as the first subordinate series in passing from Rigel to Bellatrix, and this seems to suggest that the molecules corresponding to the principal series do not survive so high a tem- perature as those which produce the lines of the first subordinate series. There is, however, no sufficient reason for regarding the three series of gas X or of helium as representing separate constituents of 58 Mr. J. Norman Lockyer. On the Appearance of the those gases, and for the present, at all events, each of the three series of helium or gas X may be taken to represent the vibrations of molecules of the same gas but of different complexities. ; The differences in the stellar behaviour of helium and gas X have been confirmed by reference to the researches of Professor Vogel* and Professor Pickering. I suggest that the time has come to give gas X a definite name. It will be remembered that I pointed out in May, 1895,f that helium was only one constituent of the gas discovered by Professor Ramsay, which he imagined to consist of helium alone, and that there was spectroscopic evidence suggesting at least one other new element associated with helium. Afterwards, in September, 1895, Professors Runge and Paschen came to the same conclusion,§ but their work still left indeterminate the number of elemental gases in the mixture. In the many comparisons of the lines I had to make in my investi- gations I soon found the inconvenience of not having a name for the gas which gave 667, 501, and other lines, and I called it gas X for laboratory use. When, therefore, Professors Runge and Paschen, who had endorsed my results, and had extended them, called upon me, I thought it right to suggest to them that, sinking all questions of priority, we should all three combine in suggesting a name for this gas, the elemental character of which we had demonstrated. This offer they declined,|| and so far as I was concerned the matter dropped. In the meantime Dr. Stoney has suggested the name “ parhelium.” But seeing that this word is already in use in another connection for a ‘‘mock-sun,” its acceptance is, I think, impossible. I propose, therefore, the word “asterium,” since it is in the stars that the behaviour of the new element has been best studied, and its appear- ance furnishes valuable evidence as to their chemistry. The probable Existence of other New Gases in the Hottest Stars. Discrimination between Gaseous and Metallic Lines.—The lines of helium, asterium, and hydrogen in the hottest stars are accompanied, as I have stated, by others which may either represent gases of a similar character, or metals at very high temperatures. 1t becomes important to consider the means at our command for distinguishing between gaseous and the metallic lines. * © Astrophysical Journal,’ 1895, vol. 2, p. 333. + ‘Annals of the Harvard College Observatory,’ vol. 28, Part I. t ‘Roy. Soc. Proc.,’ vol. 58, p. 194. § ‘Nature,’ vol. 52, p. 321. || ‘Science Progress,’ June, 1896, p. 278. Cleveite and other new Gas Lines in the Hottest Stars. ao: One possible method is this. In the nebule are found the lines of hydrogen, helium, and asterium associated with other bright lines of unknown origins; it is fair to assume that if other similar gases exist in the nebule, the other bright lines should belong to them. In the nebule all these probably exist at low temperatures, since no indication of the enhanced lines of Fe, Mg, Mn, Ti, &c., have been detected in the spectra of nebule, and on this ground we are driven to give up the old arguments in favour of the high tempera- ture of the nebule, which depended for their validity upon the presence of ‘‘chomospheric”’ lines in the spectrum. The discovery of terrestrial helium has enabled its behaviour, when rendered luminous, to be studied, and we now know that its presence in a spectrum is no proof of a very high temperature. Further, of all the lines other than hydrogen, helium, and asterium, so far discovered in the nebulae, it would appear that only a few, if any, are certainly produced by metallic vapours.* | If, then, their origins be gaseous, as opposed to metallic, we should expect to find these lines in the spectrum of those stars in which the absorption of hydrogen and the cleveite gases which are associated with them in jthe nebule is strong. At present, this method of separating the gaseous from the metallic lines in the hotter stars cannot be finally applied, for the reason that the wave-lengths of many of the nebular lines are not sufficiently accurate for the object in view. But there is little doubt that it will furnish a valuable criterion when photographs with larger dispersion become available. ) Another possible method, however, is open to us. In @ Cygni, where the enhanced metallic lines are so strongly developed, the helium lines appear very feebly, and it is only in stars at still higher temperatures that helium is strongly represented. Hence, if there are other gases which behave like helium, in stars as well as nebula, they would be intensified in passing from « Cygni through succes- sively hotter stars, while the enhanced metallic lines become feebler. Some of the principal lines which become thus intensified in passing to the hottest stars are indicated in the following table. Five of the lines given in the table approximately coincide with enhanced lines, two with lines of cadmium and three with lines of sulphur, but since in the spectrum of the former substance there are fifteen enhanced lines in the same region, and in the latter twenty- nine, the coincidences may for the present be regarded as acci- dental. It seems highly probable therefore that the lines recorded in the table represent gases which have yet to be discovered, and that the * Phil. Trans.,’ A, 1895, vol. 186, p. 76. 60 Mr. J. Norman Lockyer. On the Appearance of the Lines other than Hydrogen and Cleveite Gases which make their appearance only at Temperatures higher than a Cygni, or become intensified at higher Temperatures. Wave-length. a Cygni. Rigel. Bellatrix. é Orionis. -=-———— —_————. 3919 °2 3994 °7 4.040 °6 4069 °7 4071 °7 4075 °7 4088 °7 4094°7 4104 °8 4114 °8 4172- 4253 ° 4267" 4314 ° 4345 ° 4415 4541 ° 4566 ° 4574: 4613° 4643 4650 fel Sh alee ately Ie dbs Teele US| Gl alee 5 | | | | nono | wo ny | Lal eae! | oo | Pe lesa lett tetas lol la] eas COMDNHDWDNBHORHAGD | intensification of Feead in the hottest stars in passing from « Cygni is is a. trustworthy criterion for gaseous lines. If there are other gases which, like hydrogen, give indications of their presence at the temperature of « Cygni, or lower, the lines in the spectrum do not, like those of hydrogen, become more intense with increased stellar temperature. In such cases it does not seem likely that anything short of the actual discovery of terrestrial sources of the gases can help us to differentiate the lines belonging to them in stellar spectra from those due to metallic substances. Attempts to trace Terrestrial Sources of the New Gases—In a series of papers communicated to the Royal Society I have given an account of the attempts which I have made to find new gases by experi- ments upon minerals similar to those adopted for the extraction of helium and the associated gases from cleveite. In the last paper of that series I summarised the results which had been obtained, indi- cating that lines occurring in the spectra of gases from minerals for which no known origin could be assigned were represented in the spectra of some of the hotter stars.* From this I extract the following list of the lines thus found to * © Roy. Soc. Proe.,’ vol. 60, p. 133. Cleveite and other new Gas Lines in the Hottest Stars. 61 have probable counterparts in the hotter stars, and show the inten- | sities in different stars. Lines which have since been found to correspond with enhanced lines in the spectra of any of the sub- stances so far examined are omitted. Stellar Lines probably coincident with Lines in the Spectra of Gases from various Minerals. | a Cygni. Rigel. Bellatrix. 5 Orionis. Waveleneth | Max. = 10. | Max. — 10, | Max. = 10. | Max. = 10. 3929 °4 3961 6 4002 °9 4069 -7 4072 °2 4114°6 4309 *4 4338 ‘0 EU LLer eee (S| ce ere aml] Hl] cos | The lines thus observed in the spectra of mineral gases may be divided into two groups, the first comprising those which are strongest in « Cygni and thin out at higher temperatures, and the second those which are either absent from « Cygni or become stronger as the temperature of « Cygni is exceeded. The lines of the first group behave like the enhanced lines of the metals, and, unless the gases can be isolated, it is impossible to say whether the corresponding stellar lines are produced by gases or the coincidences are merely accidental. In the case of the second group of lines, however, the probability that the mineral gases which give them really represent new lines is much greater, since, like the lines of helium, they are most intense in the stars which we have every reason to believe to be the hottest. Attempts to Trace Series of Lines——A minute examination of Bellatrix has been made by Dr. W. J. S. Lockyer with a view of inquiring whether some of the many still unknown lines might possibly form series like those of helium and asterium. With this object the lines in the spectrum were carefully plotted, Special attention being given to the intensities of the individual lines. An examination of the residual spectrum formed by omitting all those lines the origins of which were known, showed that possibly two further series were present, but a better photograph of the star spectrum is required to settle the matter definitely. The residual lines in the spectrum of Bellairix in the existing photographs after hydrogen, helium, and asterium have been with- ppearance A WOnsthe yer Mr. J. Norman Lock 62 Cleveite and other new Gas Lines in the Hottest Stars. 63 drawn, amount to upwards of fifty, and there can be no doubt that some of them represent gases not yet discovered on the earth. It may also be stated that these gases behave differently as regards their range of visibility through stars of varying temperatures. The accompanying map shows the principal lines in the spectrum of Bellatrix, and indicates those which are due to hydrogen, helium, and asterium. The two probable new series which have been found by Dr. Lockyer are also shown. The New Gas in ¢ Puppis. Professor E. C. Pickering has recently announced* that the spectrum of ¢ Puppis contains a new series of lines, which he at first supposed to be due to some new element. In a second communica- tion to the same journal,+ he pointed out that he had reason to suppose that this new series was in some way connected with. hydrogen, since he found that the observed lines occupied the same positions as those computed from the same formula and constants, from which the ordinary series of hydrogen was calculated, but using odd values of 7 instead of even values. In the following table I have brought together all the lines published by Professor Pickering as belonging to the spectrum of this star, ranging them in different columns for greater clearness :-— Lines in the Spectrum of ¢ Puppis. Hydrogen. Other lines with origins. Old Seri | New Seri eries ew Series 5 (dark). (dank); Dark. Bright. 3798 °1 3783 *4. 3933 Ca 4698 unknown 3835 °5 3815 °9 4472 He 5652, 3889-1 3858 °6 4505 unknown 3970 °2 3924.°8 4620 a 4101 °8 4026°8 4633 i 4340 ‘7 4.200 ° 4. 4688 BA 4861 °5 4544 *O | In his first communication, Professor Pickering mentions lines at 4698, 4652, 4620, and 4505, but he does not refer to the first three in his second paper. The line 4505 was at first taken to be one of the components of the new series, but this seems to have been subse- * © Astrophysical Journal,’ vol. 4, p. 369. + ‘Astrophysical Journal,’ vol. 5, p. 95. 64 Mr. J. Norman Lockyer. On the Appearance of the quently superseded by the employment of the line about 4544, which agrees better both as regards intensity and the calculated position 45436. | The question then arises, what relation does the spectrum of this star bear to those of other stars. of high temperature ? A comparison of the lines recorded with those in the spectrum of Bellatrix shows that, with the exception of the new series and the lines at 4698, 4688, and 4505, many lines are common, as is indicated in the following table :— Comparison of Spectrum of ¢ Puppis with that of Bellatrix. Lines in Z Puppis Probable coincident | (Pickering). lines in Bellatrix. Origins. 3763°4 —— H new series (n = 21). 37981 3/981 Ho. 3815°9 == H new series (n = 19). 3835°5 3835°5 Hy. 3858°6 — H new series (x = 17). 3889'1 3889°1 ie 3924°8 —. Hs new series (7 = 15). 3933°0 39330 K. 3970°2 39711 He. | 4026°8 — H new series (7 = 18). 4101°8 4101-0 He. 4200°4 oe Hi new series (7 = 11). 4340: 7 4340°7 Hy. 447 2:0 4471-0 Helium. 4.505'0 = Unknown. 4.540 °0 4541-0 Unknown. 45440 = H new series (x = 9). 4620-0 4620:0 Unknown. 4633°0 4629-0 Unknown. 4652°0 4.650°0 Unknown. 4688°0 — Unknown. 4698°0 = Unknown. 4861°5 4861-0 Hp. From the above it will be gathered that the only really marked difference between ¢ Puppis and Bellatrix is the presence of this new series of Jines in the spectrum of the former. As I have only at my command the published accounts of Professor Pickering and not an original photograph of this star, a more detailed comparison of the spectra cannot be made, but there seems to be evidence which points to a higher temperature for the star than that of Bellatrix. Professors Pickering and Kayser both concede that this new form Cleveite and other new Gas Lines in the Hottest Stars. 65 of hydrogen is due most probably to a high temperature, and Pro- fessor Kayser expressly states that “this series has never been observed before, and can perhaps be explained by insufficient tem- perature in our Geissler tubes and most of the stars.”* I pointed out in my former paper that this new series and the one previously known are probably of the subordinate type, and that the principal series is still unrecognised,} although some of the “unknown ” lines in stars may possibly belong to it. On the supposition that the new series of probable hydrogen lines in ¢ Puppis represents the effect of a transcendental temperature, an attempt has been made to produce this spectrum in the laboratory. In the high-tension spark in hydrogen at atmospheric pressure the ordinary series of hydrogen lines is very broad, and none of the new series have so far been detected. ‘The use of the spark with large jars in vacuum tubes results in the partial fusion of the glass and the appearance of lines which have been traced to silicium, while the new series has not yet been observed. Final Result as to Temperature. In the preliminary attempt to determine which are the hottest stars, the following facts and deductions :iave been considered :-— 1. With increasing temperature hydrogen is first visible, then helium and asterium appear nearly together, and finally unknown lines at \A 4088°7 and 4650°9 make their appearance. 2. The chief helium lines in the region covered by the photographs become much thicker after the « Cygni stage has been passed, and are practically of equal thicknesses in the stars Bellatrix, Spica, 6 Orionis, ¢ Orionis, « Orionis, and # Persei, after which a sudden diminution in intensity takes place. These lines give us no critevion for the hottest star of the series. 3. With regard to the chief lines of asterium, namely, 4008-7 and 4388°1, in the region under investigation, these both rise to a very decided maximum in Bellatrix, diminishing afterwards in intensity less rapidly than they increased. The great development of asterium after the lines of helium have reached a considerable thickness sug- gests a higher temperature for Bellatrix than the neighbouring stars in the series. | 4, As asterium begins to decrease in intensity, the two unknown lines before referred to at AA 4088°7 and 4650°9 commence to brighten, reaching a maximum at e Orionis, in which the hydrogen lines are still at a maximum, but asterium has considerably decreased. Only a trace, if -any, of these lines can be found in Bellatrix. If * © Astrophysical Journal,’ vol. 5, p. 96. + ‘Roy. Soc. Proc.,’ vol. 61, p. 195. yoL. LX. B 66 Mr. J. Norman Lockyer. On the Appearance of the these lines when at maximum are indications of a higher tempera- ture than those of asterium, then, since the hydrogen and helinm lines are also at a maximum, ¢« Orionis on this assumption would be the hottest star of the series. With our present data it is, however, difficult to state with certainty whether the principal series of helium or of asterium makes its appearance first. There seem, however, to be indications which suggest that asterium is a somewhai later development. In summing up I may say that « Orionis may be considered the hottest star from the behaviour of the lines at 4481°3, 4088-7, 4650°9, while 4008°7, 4267°6, and 4388°1 favour the star Bellatrix in this respect. The helium lines (40263 and 4471-6) have practically the same intensity in both stars, or at any rate there is not sufficient difference to serve as a criterion. General Conclusions. 1. The order of temperature of stars at and near the apex of the temperature curve can only be determined by reference to unknown lines, since the enhanced lines of iron are absent, and those of mag- nesium and calcium are exceedingly feeble, while the lines of the known gases show no very marked variations. 2. The varying appearances of the lines of the cleveite gases indi- cate, as laboratory work has done, that helium and gas X are distinct substances, but there is not yet sufficient evidence for regarding the constituent series as belonging to separate substances. It is therefore considered that gas X should be definitely named, and the name ‘‘asterium”’ is suggested. 3. There are two methods open to us for discriminating between gaseous and metallic lines of still unknown origin in the spectra of the hottest stars. (a) Gaseous lines like those of helium and hydro- gen will be common to nebule and the hottest stars. (6) Metallic lines like those of iron, magnesium, and calcium will thin out at increased stellar temperature, while gaseous lines will become inten- sified. 4, Several unknown lines in the spectra of the hottest stars are thus shown to be most probably of gaseous origin. 5. Attempts to trace terrestrial sources of these stellar gases have resulted in the detection of lines which probably coincide with lines in the spectra of the hottest stars. 6. On the supposition that these stellar gases are more or less allied to helium and asterium, since they have their maximum inten- sity in the same stars, attempts have been made to trace “ series”’ of lines in the spectra. In the case of Bellatrix two probable series have already been recognised, | . ERRATUM. Vol. 61, p. 410. In the memoir read June 3, 1897, entitled “The average Contribution of each several Ancestor to the total Heritage of the Offspring” the words sire and dam have been accidentally transposed in Table II, and consequently, in the deduction therefrom at the bottom of page 404, the latter should be that, in the present case, the sire is more potent in transmitting colour than the dam, in the ratio of 6 to 5. This error does not affect the general conclusions of the memoir, because the ratio of 6 to 5 was treated as an insignificant disproportion, and the two sexes were dealt with on equal terms. FRANCIS GALTON. Cleveite and other new Gas Lines in the Hottest Stars. 67 7. The new series of probable hydrogen lines in ¢ Puppis most likely represents the effect of a transcendental temperature. This and the well-known series are in all probability of the subordinate type, the lines of the principal series not yet having heen identified in stars. | 8. There is evidence which points to a higher temperature for ¢€ Pappis than for Bellatrix. 9. The behaviour of certain lines suggests that Bellatrix may be taken as a type of the hottest stars, while the behaviour of others seems to indicate that « Orionis should be regarded as a star of the very highest temperature, exception being made of ¢ Puppis. There are not yet sufficient data to enable a final statement to be made. “A Maya Calendar Inscription, interpreted by Goodman's Tables.” By ALFRED P. MaupsLay. Communicated by F. DucaANE GODMAN, F.R.S. Received April 2,—Read June 17, 1897. [ Introductory Note. Our knowledge of the Maya Calendar is chiefly derived from the writings (A.D. 1566) of Diego de Landa, Bishop of Yucatan, who not only gave some account of the divisions of time in use among the Mayas, but also copied, somewhat roughly, in his manuscript the signs employed to represent the eighteen named months, and the twenty named days into which each month was divided. Landa’s statements are, however, by no means clear, and there has been much discussion both as to their correctness in themselves and as to the interpretation which has been given to them ; moreover, it has been found difficult in some instances to identify the day and month signs given by him with those used in the Dresden Codex and the few Maya manuscripts which have been preserved, and still more difficult to identify them with the signs used in the carved inscrip- tions. In the accompanying paper an examination is made of a recently discovered inscription, by the aid of calendar tables prepared by Mr. J. T. Goodman, and published with an explanatory essay in the ‘Biologia Centrali-Americana.’ These tables consist of a chrono- logical and an annual calendar. The chronological calendar is based on the Ahau, a period of 360 days, and is divided thus :— CORE ie oe ane 1 chuen . Lerchuens =". 1 ahau* (360 days) * It is unfortunate that the ahau, or period of 860 days, bears the same name as one of the twenty days of the Maya month, and that the chwen, or twenty-day period, bears the name of another day of the month. VOL, LXII. G 68 Mr. A. P. Maudslay. A Maya Calendar Inscription 20 ahawe ees fs 1 katun 20 etn. 6 25°s.. 3 1 cycle ‘TS eycles!........2 “Tieresineygete 73 great cycles .... 1 grand era The annual calendar is divided into eighteen named months, each consisting of twenty named days, and one short month (named Uayeb) of five days. The twenty named days of the month are numbered continuously from 1 to 13, so thatif the first-named day of the month has the number 1 attached to it, the last-named day of the month will be numbered 7 (13+7 = 20), and the first day of the next month will be num- ‘bered 8, and so on. There are fifty-two annual calendars in a calendar round, and at the end of the 52nd year the series is repeated. All the dates and reckonings found on the monuments which can ‘be made out by the aid of these tables are expressed in Ahaus, Katuns, &c., and not in years; but Mr. Goodman maintains that the true year was known to the Mayas, and that it is by the concurrent use of the chronological and annual tables that the dates carved on the monuments can be properly located in the Maya Calendar. All the dates which have as yet come under notice fall within the three Great Cycles, numbered by Mr. Goodman the 53rd, 54th, and 55th. The following extract from an article in ‘ Nature’ (July 8, 1897) ives a good example of the manner in which a date is expressed :— ‘“‘T called attention, some years ago, to the fact that the greater number of the carved inscriptions commenced with easily recognised series of glyphs with numerals or faces attached to them, which I ealled the Initial Series. Mr. Goodman now shows that the Initial Series expresses a date thus :— (1) The Great Cycle sign. (2) The Cycle. (8) The Katun. (4) The Ahau. (5) The Chuen. (6) The Day. (7) The named day. (8) The named month. interpreted by Goodman’s Tables. 69 As has been long known, each bar counts as five, and each dot as a unit. (The roundish marks wnder the glyphs are not part of the numerical series. ) “The sigus in front of the Ahau, Chuen, and Day signs denote a ‘full count’ of those periods. The date thus reads :— WUE wd Sede a ld ce 3 great cycle. 2) Te) de deg Bites cycle. BCS aaa sd 5 oe ges katun. (45) “all count’? ....5. 6% ahaus. eae ull comnts ys Yo os chuens. me, Hulleonmt’ .5¢ oc... days. (7.) 4 Ahau (day). _ (8.) 13 Yax (month). *« A reference to Mr. Goodmaun’s chronological calendar shows that the 15th Katun of the 9th Cycle of the 54th Great Cycle commences with the day 4 Ahau, the 13th day of the month Yax, the date which is here given in the inscription. The combination 4 Ahau 13 Yax ean only occur once in a period of fifty-two years. “ One of Mr. Goodman’s diseoveries is the system on which the Mayas numbered the different series of time divisions. For instance, the twenty Ahaus are not numbered 1, 2, 3, &c., up to 20, but they were numbered 20, 1 2, 3, &., to 19. “Tf we should nowadays wish to use a similar notation, we should probably number the series 0, 1, 2, &e., 19; but it seems as though the Mayas, having no sign for 0, wrote the sign for 20 or a ‘full count’ of Ahaus in the first place. “The eighteen Chuens are in like manner numbered 18, 1, 2, 3, &c., to 17, the same sign being used for a ‘full count’ of Chuens as is used for a ‘full count’ of Ahaus. “ Asa ‘full count’ of days (twenty) is a Chuen, a ‘full count’ of Chuens (eighteen) is an Ahau, and a ‘full count’ of Ahaus (twenty) isa Katun. The foregoing inscription inay be read thus :— “The 15th Katun of the 9th Cycle with no odd Ahaus, Chuens, - or days added, begins with 4 Anau 13 Yax. “ Had the date been one including a specified number of Ahaus, Chuens, or Days, we should have had to make use of the annual calendar. “The faces so frequently met with in the inscriptions in connexion with Cycle, Katun, and other signs for time periods are shown by Mr. Goodman to be in reality numerals, and the whole series of numeric faces from 1 to 20 has been determined in some cases with certainty, and in others with a fair degree of probability.”—August 5, 1897. | In the month of February, when the last pages of Mr. Goodman’s Re egies Mr. A. P. Maudslay. A Maya Calendar Inscription 70 ee oy ry) ; ‘al (CG (=I y) ) v 4 — lige Z NWS S31 2) | cman = y = ) — o> ( BS SSS a SS ee on Maya Inscription from Piedras Negras. (The glyphs are read downwards in double columns from left to right.) interpreted by Goodman’s Tables. 71 essay (published in the Archeological Section of the ‘ Biologia Centrali-Americana’) were issuing from the press, I received from Mr. Teobert Maler a number of photographs of sculptures and inscriptions which he had recently discovered in Yucatan and the country to the south of it as far as the banks of the River Usumacinta. ; One of these inscriptions from Piedras Negras on the Usumacinta is in a good state of preservation, and a drawing made by Miss Annie Hunter from the photographic print is here reproduced (p. 70). As Mr. Goodman has never seen this inscription, an examination of it with the help of his notes and calendar tables will be a fair test of their value. The following signs are figured in Mr. Goodman’s essay, and will be found to agree fairly well with those in the inscription. Cycle Katun = 20 Ahaus Ahau = 18 Chuens. A Jf ee 20 Days. 72 Mr. A. P. Maudslay. A Maya Calendar Inseription Signs for the named days— Cimi. Cib. Ymiz. Ahau. Signs for the named months— YC. Yaxkin. Kankin. The glyph A 1 is the initial glyph indicating the Great Cycle. It has more the appearance of the sign for the 53rd than for that of the 54th Great Cycle, but the signs for the different Great Cycles are still in need of elucidation, and the subsequent reckoning shows clearly that the dates fall within the table given by Mr. Goodman as that of the 54th Great Cycle. The next glyph B1 is the Cycle sign with the numeral 9 in front of it (one bar = 5 and four dots = 4). A 2 is the Katun sign with the numeral 12 in front of it (two bars = 10, and two dots = 2; the hollow curve between the two round dots is merely used to fill wp the space, and does not count). B 2 is the Ahau sign with the numeral 2. Turning to the tables of Mr. Goodman’s chronological calendar, of which an extract showing the 10th to the 4th Katuns of the 9th Cycle is here given, we find that the first day of the— 2nd Ahau, 12th Katun, 9th Cycle, 54th Great Cyele, falls on the day 2 Ahan, the 18th day of the month Xul (which is underlined in the table). This is as far as the chronological calendar ean guide us. We have next to find the position of this date in the anuual calendar. The date can only occur onee in the fifty-two years which constitute a calendar round, and an examination of the tables shows that it falls in the first year of the calendar round (where it is marked with a square). interpreted by Goodman’s Tables. SYRXNHNvwroonwoa X - ‘ney aT} JO ‘ON XB8X | 8ST 8 ie ¢g oI c< 8 ge — €1 vs 0e7 | TI II 66 e A = 8 9 a: €1 OL eal Oi ook 66 @ Gc = 8 6 * eI SI oBN | 8T P 6¢ e 8 e 8 rl eI e ulyuey | gt fs " g Il (<4 Q gS usu | St 9 ers ee B | ie ® Si ae °o a co ee ee ° B a p ) © i =A a : > YT “YyUOUL JO oUItNy ‘Yguout ayy Fo Aug > NS @ By ep oy JO “ON s eI oI Oise eal € (a9 re 4 se 8 II os gI1 3 diz | 81 9 4 g ro} § “ce g 7 = eA G 2407 | 8I 6 a € rs 8 5 ad EL 8 09277, | 8I aI (73 cla Pe z eT Il NX | 81 Fe g 9 uUIyxex | 8 OL g<) oes B | o° ® o. eo ° ot ct 2 lee os ae So 5 = 2 | | 4 ar apoAD YIUIN 66 UL YX 6c 6¢ “Y}UOUL JO 9UIe NT apoAK yeoayy yyanog-AG FLT bb éT SI : ; OC 9 00 09 00 C2 GO cd GD 20 1 OD qjuou oy fo Leq Be here ey aa OP Oe) Be eS Ley kep oy Jo ‘on Os DRAMA OONAKEORDYWVWOAWH rt re S| we ie ie fs) OF YO} st 2, Sb - g Il LL se 8 3S OL SI 9 Gh OVI | «ST OL Val g T EL Hs 8 ON ob - et 6 th uryuey | St OT OL H g 57 6 ¢é g 8 8 = SI SI he Uen JL ST +S 9 it4 g 2, G 3. 8 II Y “1 S £8 xg | 81 9 ra g OL T qeivy | 8 T 06 Z Pee 5 B | Sale. s rs 2 pa > B (ae) (a) (a9) g : a = ct, 5 a is) a cS s Of *"sunqey 91 JO ‘SON NOH FIDO k OD A Maya Calendar Inscription SYNH WH SCRBDDA Sy) Mr. A. P. Maudslay. 74 Penne nee eect n ee ee eee TUE Preteen eee eter eeee eee ney Freee ee ene eee eee tees guna eee eo Tey ta Peete eee eee eee seen ee UBqES N re — rt — a No yi TF 1DOR OD oO e oo wince & wD x) rei qteeseewerpecdee sees ones OTT ooee "ee eo ee ee oe ee ewe ee wd oe XI rN OT ORES. WR es) 6 Oe De Nake 8) eee) ee ee: WO ull saanlipe pana a eRe Be ot ee or Da hoo ain: etter oe ccs ti BEE ge ot eg pet TG) es anaes rire a oe ee ee Oe oe oe oe oe ot Oe Oe I2Q1V ae einai PAL Le), || HAMM OoOK ORO N ~ NOHAMBWAOKRDRO core re aN =o HADYED)OND AS e . ° . . . . ° e . : . ° . . . . . . NMOMAN O19 OM OG => = ANMFAWMWOPDHOANMAN DT OY Se OMm~ODOHDOnN ee se ‘shop ay2 fo SawDAT OO Ss OC OO OO OO | ———_ | |S | L | sere eeee STAUOUL OTF JO SOULE NT “IV9 X 4ST ‘AVPUGTVO [BUUUY olVypouy interpreted by Goodman’s Tables. 6: The next glyph in the inscription A 3 is the Chuen sign with the sign which signifies a “full count” of Chuens, in front af it. Asa full count of Chuens is 18 and equals 1 Ahau, and as the number of Ahaus has already been recorded, the glyph A 3 means that no odd Chuens are to be added to the date already expressed. The glyph B 3 is the sign for a day (of twenty-four hours) pre- ceded by the numeral 16. Turning to the first year of the annual calendar, we now add these 16 days to 2 Ahauw 18 Xul, the date already arrived at, and it will be found to bring us to 5 Cib 14 Yawxkin (marked with a circle). That this reckoning is correct is shown by the inscription itself where the result is expressed; A 4 being 5 Cvb, and B7 14 Yavrkin. The six glyphs in the inscription intermediate between the sign of the day Cib, and the sign of the month Yaxkin, have not yet been thoroughly deciphered, but there is reason to suppose that they contain a parallel reckoning differently expressed. The next three glyphs are undeciphered; then comes another reckoning:— _ - C 1 is the Chuen sign with the numeral 10 (two bars = 10) above it, and a “full count” sign at the side. Whether the 10 applies to the Chuens or days can only be determined by experiment, and such experiment in this case shows that the reckoning intended to be expressed is 10 Chuens and a “full count” of days, that is for prac- tical purposes 10 Chuens only, for as in the last reckoning when the full count of Chuens was expressed in the Ahaus, so here the full count of days is expressed in the Chuens. The next glyph D 1 is an Ahau sign, preceded by the numeral 12. This gives us— 12 Ahaus (12 x 360) = 4820 days. 10 Chuens (10x 20) = 200 ,, 4520 days. 4380.5. == 12 years. —— 140 days. adding 4520 days, or 12 years and 140 days, to the date 5 Cib 14 Kankin, it brings us to the date 1 Cib 14 Kankin i in the thirteenth year of the annual calendar. Turning to the inscription we find at C 2 (passing over the first half of the glyph), 1 (ib followed by (the first half of D 2) 14 Kankin, the date at which we have already arrived by com- putation. upton A Maya Calendar Inser Mr. A. P. Maudslay. 76 Sater et Pe rere or ee ee ca a Ole eee Be ea, Tee OT | O° |e eres ee eee en pel ee See Bee | TT. en eee ee Rah he |e | Be ene) Qe |e he eee nae ie Oe Y eee er | 3 Lhe? | OF 1s 6 6 |8 I L €1 | 9 ot | & TALE en | aeons ee ee G TI P OL g 6 Zz 8 I L eT 9 ZL G tt C7 OL g le ac ee ee et ef Ue ie Pir | OL) & 6/4 8 if L €— | 9 or | TL iP OT 18 6 Cie | EP GP ae ere eee ee g 6 Zz 8 @ 2 eT 9 Gi G ie P OL g 6 Zz 8 I eve eoeb eed eee oe ee ee 08 OF Be dE qi ee Zz 8 T ZL eT 9 21 G IL — or g 6 Zz 8 T hi el Bar See ee eee eee eL ee T L eT 9 FAIL G IL C7 OL g 6 Zz 8 | L el 9 Ark ae a a Set ae Na > | oe ei 9 21 G TL i i OL g 6 KA 8 I M4 eT Gg ZI G Gi pen on te” Se ued Pee ape PS Il | Pv 0) ae 3 6 Zz 8 T L €1|9 ZL |g Il | & OT ok oe Eee aah a Si gs ea, ee TL P OL ¢e 6 be, 8 I Ly eT 9 OL g tL P OL e 6 ae abpanmmwontgace 2h °° uonyy ee OL ¢ 6 Zz 8 T is eT 9 ZI G il Si OT | g 6 z 8 EES ee eee 20 ate ES. 6 8 I L eT | 9 mn eG Il | OL |e 6 Zz 8 r J [rete dee ceceonseomaets ony oe 8 I L eT 9 PALL G II P OT e 6 Z 8 I Ny, eT 9 seo eo ee oe ee oe oe de OH Ow Oe qeuley eo Ms, EI 9 SL Cc IL i 4 OL ( 6 GZ 8 T L &T 9 ZL G eooe eevee Oe ee ew ee oe Fs OO yluvyy STHOn 260 | Ga Oe OR eRe (eee Oe ete ee SI Oral SE se | Tha Py aie ae Fleece ih OTe inGee| ee Gaal y. OBI 4 Oe enor it py |} Ope eG fet en Pee: ee eee) IL | ¥ OL g 6 ‘G 8 IT Us 1 €T 9 ZL G II P OL e 6 Z Ce REC an TITS SS LIEW 0 ie} ieee ie Pet io ein Se tere eo sei ak WOne es 1G) (Se) Be Le ee eda evegeee e ey 2 Oe acre VALE PO dete Ge aes @ AT |. SLs) ee eee | ‘shop ay2 fo saumn\T Q N rd eele/FlElE/lelriflelere/eleisleiFis ©, e 8, . 5 E: : s 5 E ic) t ¢ beceveda sqjUOU oy} JO SoUeNT ‘dO X UIET SYN H WHS WDD Sy) (7 interpreted by Goodman’s Tables. 6 = SpA Lp | op) GL as Va ee &b fe ol ee bh ae OL ey SYWHN WOH OwWS &N Orme A oO pea NOAH 19 OM DODO AN ec mA SF 1D OM OO 0 6 8 4 | 81 9 | 8I Gg iar vy | OT eo) 2 18 ne sI | 9 rf Il | ¥ 0) | = n| 8 =} | bet S|) et |}-o | 11 9 |2r|$ || | ot ¢ |i11|% | or| @ {6 » |orjs |6 |s |8 o.1 6c Oe | ke Sol Ste eo ery {14 /8tl9 | erie el | or elie | it) s et |¢ |it\* | ot] ¢ fii 4101) € A |e OUT & 16 | 201 ela 16 eo tk es Bola |Z et | oie het 60) O 1 BB | EL 9 |2r|¢ | It|* | ot ¢ |it|/* |orle |6 eee reese @ 1618 1 eo lag g-\@ |i |e ler te Tild | STi oo) erie N 2 |; Seah Se 5] SGEe Ss adler ae "89 X Y9GE ——- | ————— | —__ | — | | | | | | | | | | ‘| INOW 2D OFr-OOS QR oO are W190 OY DDO re ANTM OMODHS OnN co es on DF OONDRAONAMD HAG (FH) DONDE oe I oo Dn Se e@e@e78 0802020020 828 ©0806 O08 O86 uenyyO eoeee ee ee eoee eee eeee e@eoee e202 e800 026 © 2 © 8 © 90 eeee eevee ee ee ve on NAL ee ee 868 ©0208 60 e ee qeurery eevee ©G ee ee @e Oe YIU fy ee eeeoee | OSA) eeeenVreeoeee + @ uBsyoo1igg eoeesveeoe eevee © oe ee UvVy oe eb: aca Cn: 6 Ne emeeapdbe deine Ao ONT T SES a Oe eae ee Cai SEeeree oon ae eae ere OLE) ote eee SQ CUaZiT pean tee te ee soe UL CLG @) a gral og Pa CNR SS) eoeeeee ee ee ee Oe oe Udy eeenave@ee ee eveeee ee oe XI ee@eeecesee eevee ee 88 nog eooee ee ee oe ee ee oe qn ‘shop ay fo samo rT "sere suquOU OY} JO SOUB NT SeFeAaranM Oo nw Dd ist) A Maya Calendar Inscription 78 Mr. A. P. Maudslay. 6L ee ZL Se ae! aS a GL \ > | & 7 aa Maa Lal e Bae Ce a OE by at SI aS Poe = Hak MDM TNO 20 cod GYGE CN? STO KO a = (a! La HAM WwW OnRDRO HFiVTO~- ODAOnN SM Aaa ee eeee ee ee oe oe oore *e 08 @ oe or ee © e@ © eo wo oo et oo ao e@e e8 © eo eo ee ee Pee an 19 606) aRyooyD eee Uuey Steer ys were eeee yy eee eo. °° SRT = ney *- owned qeuazy * ueqeg eecoe en fe) °-* TON eoeo ee XI eee us g eee aa * uenyg Ye) * onjny * qeuLeryT + yluR yy ‘shop ay7 fo sawmvoyy ee ae ae ee a ese 2 eg | er 8 lars pat iy be Pope pele (tle |e por] |e pes Of Se | ap |r | UE | ee Or) ge gs | Be [sees ae ei) Ob eee ee. ae ee | te te COO, 8) 6-9) a) ge ps Orie Ve 6s fe 2 a here )er | se Gare ie et ol ee ee he Pt er) ere ere | op | OE ltt Pee ane | ie Vere fe. | i OF ae 2 hr he | OT @ Weare (ey its" Gel Me Ol Fe leq |e ie ih ide pt tet ie Ol Sh Ge) 218 ie een bg itt ee gk eS | Eee eenelar eget ts: Bet el eee. lg ee Pin ee ee Bee hee oe Ge es | ely | OR pe eet ee ae ity One i Gmeaka pee cL i@e li |h | Orie |e |e \ emer it M eye Or ees ice Pe eh ie rer ee Ol oe oe acer ely ere ogi k it M1 Ol) eBlH| Hl S/SINi| als ee ee eae ee |S = METIS Vltnere reeeeeee strom eq} JO sou Ny HO nDD yer oo Sa TQ >} RESUS) Fs INS PCOS S Q interpreted by Goodman’s Tables. 79 Passing over the next three glyphs we arrive at another reckoning, D 4 gives 10 days, 11 Chuens, 1 Ahan, and the first half of C 5 gives 1 Katun. Ve eee. 200 days Pea ee as so 22k D605 Pi @huens (11x20) ....2 °° 220° 3, OODTAYS! Jose css ce cae e LO: 7790 days. (665° ,,. = 21 years. 125 days. Adding 7790 days or 21 years and 125 days to the previous date, 1 Cid 14 Kankin, it will bring us to 4 Cimi 14 Uo in the thirty-fifth year of the annual calendar, and we find this date expressed in the inscription in the glyphs D 5 and C 6. Passing over the next three glyphs we arrive at another reckoning (EH 1), 3 Ahaus, 8 Chuens, 15 days :— a Ahawas) . st sisie: - . 1080 days & Ohuens: +22}... 5 EGO ti, PDA dt. oe bis ate hy. 35, 1255 days. 1095 ,, = 3’years. 160 days. Adding 3 years and 160 days to the last date, 4 Cim7z 14 Uo, brings us to 11 Ymia 14 Yaw in the thirty-eighth year of the annual calen- dar; this is the date we find expressed in the glyphs E 2 and F 2 of the inscription. It is true that in the sign in the glyph E 2 is not the sign usually employed for the day Ymix, but that it is a day sign we know from the fact that it is included in a cartouche, and I am inclined to think that the more usual Ymix sign (something like an open hand with the fingers extended) was inclosed in the oval on the top of the grotesque head, but it is too much worn for identification. Passing over seven glyphs, the next reckoning occurs at F 6, which gives :— A CHGCMS pose 0 s:50-0's 80 days WON Day ses Pe. LUNN. Os HO b,, 99 days. Adding 99 days to the last date, 11 Ymix 14 Yaz, brings us to 6 Ahau 13 Muan iv the same year, and we find this date expressed in F 7and F 8. 89 Dr. A. D. Waller. Influence of Acids and Alkalis The last glyph in the inscription is a Katun sign with the numeral 14 above it, and a sign for ‘‘ beginning” in front of it, and indicates that the last date is the beginning of a 14th Katun. If we turn to the table for the 9th Cycle of the 54th Great Cycle, from which we started, it will be seen that the 14th Katun of that cycle does commence with the date 6 Ahau 13 Muan. It is simply impossible that the identity of the dates expressed in the inscription with those to which the computations have guided us can throughout be fortuitous. Very nearly half of the forty-eight glyphs in the inscription have been accounted for, and I have no doubt that when the inscription passes under Mr Goodman’s scrutiny he will be able to give us much information about the remaining glyphs which I have passed over as undeciphered. Tt can, I think, therefore, be fairly claimed for Mr. Goodman that his researches have raised the veil of mystery which has for so long hung over the carved hieroglyphic writing of the Mayas. “Influence of Acids and Alkalis upon the Electrotonic Cur- rents of Medullated Nerve.” By Augustus D. WALLER, M.D., F.R.S. Received June 10,—Read June 17, 1897. A. The Effect of Acids and of Alkalis. Considering that electrotonic currents are characteristic of living medullated nerve, that such currents are due to electrolytic polarisa- tion, and that such electrolysis must primarily consist in a libera- tion of electronegative principles (oxygen, acid, &c.) at the anode, and of electropositive principles (hydrogen, base, &c.) at the kathode, the first and most obvious test to be made is to examine com- paratively the action of acids and bases upon anelectrotonice and katelectrotonic currents. On the supposition that a medullated nerve-fibre is composed of two different electrolytes, white fatty sheath and grey proteid axis, and that electrolytic polarisation is aroused at the interface of separation between these two electrolytes, we may expect to find, as the characteristic acidic effect, diminution of A and increase of K, and as the characteristic basic effect, increase of A and diminution of K. This expectation is in the main substantiated by experiment, although owing to the somewhat narrow range of concentration within which moderate effects are produced, it is not common to obtain effects in both of the two opposite directions in a single experiment. The reagent may be too weak, in which case neither A nor K are altered, or it may be too strong, in which case both A upon the Electrotonie Currents of Medullated Nerve. 81 aa pn en be DD ESS EES SSiterce a > gf Leathe e SSS ES SRS CG KGAA AAA RAGS SEER DEVE NERLUCERRLARRARNEN = Yer __-¢ VAT REEL oe = CEE is eg == SESS See eee eee ee Py Sy Pe => KK —. : L Te. veal. iM \S n - a / 1, 5 6 55 AA eS a” LN r) 4) i xy ifefete) ie" iste 7? wos, Ls Fie. 3.—Effect of sulphuric acid, N/S upon A and K (2359). and K are rapidly and equally abolished. Plates 9358 and 2359 illustrate this point, the former exhibiting the defective action of 82 Dr. A. D. Waller. Influence of Acids and Alkalis a base below optimum strength, the latter exhibiting the excessive action of an acid above optimum strength. Partly for this reason, and partly in order to eliminate the re- sistance factor, results are formulated in terms of the relative magni- tude of the quotient A/K as well as in terms of the absolute magnitudes of A and K. This point is illustrated by plate 2410. Fie. 4.—Oxalic acid N/20 (2410). Method.—The disposition of the object of experiment is 1m accord- ance with the diagram, and the galvanometer (dead-beat) is arranged Fie. 5.—Diagram of apparatus. The excised nerve rests upon two pairs of un- polarisable electrodes, pp’ leading in the polarising current, ee’ leading out the extrapolar or electrotonic current. to give a continuous record (lasting usually from 30 to 60 minutes) as described in a previous communication (Croonian Lecture, 1896, ‘Phil, Mranis!, 41. 1897), upon the Electrotonic Currents of Medullated Nerve. 83 In the earlier observations series of anelectrotonic and of katelec- trotonic currents were separately recorded. In the later observa- tions the A and K currents were taken at alternate minutes by means of a rotating reverser in the polarising circuit. In the finished records A currents read upwards, and K currents read downwards. Resulis—The characteristic results of acid and base upon the anodic and kathodic currents respectively are summarised in the following four observations (Plates 2360, 2412, 2429, 2432). la mos aes Fic. 7.—Action of ammonia vapour on A and K (2432). They illustrate the rule that : Acidification diminishes the quotient A/K. Basification increases the quotient A/K. A diminution of the quotient A/K may be by diminution of A or by increase of K. In plate 2412 it is mainly by diminution of A. In this case the augmentation of K is comparatively small. The record in fact approaches towards the type of plate 2359. In other VOL. LX. H 84 Dr. A. D. Waller. Influence of Acids and Alkalis Fie. 9.—Effect of a weak alkaline bath (KOH N/20 or 0-285 per 100) upon A and K (2360). experiments (e.g., in plates 2429 and 2424) the diminution of A/K is principally due to increase of K. An increase of the quotient A/K may be by increase of A or by diminution of K. In plate 2560, and in most of my other experi- ments, it is mainly by diminution of K. The two plates illustrate the points that acid affects A more than it does K, and that base affects K more than it does A. As mentioned above, acid above optimum strength causes a diminu- tion of K; we may, therefore, say that weak acid causes augmented K, and stronger acid diminished K. On testing carefully with very weak acids we shall find that ata strength below the optimum (giving diminished A and increased K, e.g., Plate 2429) the “very weak” acid causes augmented A. We may, therefore, say that very weak acid causes angmented A and rather stronger acid diminished A. And in general summary of the action of acid from minimal to maximal effective we may state that :— upon the Electrotonie Currents of Medullated Nerve. 85 _ (1) The weakest acid gives increased A. (2) Slightly stronger acid gives diminished A and increased K. (3) Still stronger acid gives diminished K. These statements are the outcome of a considerable number of observations, and one may hardly hope to verify the progressive action of acid from minimal to maximal in a single observation with a single acid. Nor is it easy to give numbers in lieu of the indefinite qualifications “ weak” and “strong.” This much may, however, be said to give an idea of the order of magnitudes dealt with. The second degree of change may be expected in consequence of bathing the nerve for one minute in an acid solution of a strength between N/20 and N/10. The free passage of “much” CQO, into the nerve-chamber usually affects the second degree of change in its most typical form. A small amount of COQO:, e.g., a few pufts of expired air, will more probably affect the first degree of change. A bath of one minute’s duration in a N/5 solution of mineral acid will almost certainly affect the third degree of change. A diminution of K by CO, is rare (e.g., 2363). B. The Effect of Carbonic Acid and of Tetanrsation. I have given particular attention to the action of carbon dioxide and of tetanisation upon the A:and K currents, in prosecution of observations already reported concerning the action-currents of nerve* and the influence of temperature upon the A and K currents. The usual and typical effects of carbonic acid are of the charac- teristic acidic type, consisting in a diminution of A and an augmenta- tion of K. i Fig. 10.—Action of CO, upon A and K (2422). * Phil. Trans.,’ B, 1897, p. 1. 86 Dr. A. D. Waller. Influence of Acids and Alkalis Less commonly, and by a slighter degree of action of CO,, the A current may be increased, while as the most pronounced degree of action of CO, the K current may be diminished. In order of gravity the effects are: 1. Augmentation of A. 9 ee of A. ' | Augmentation of K. 3. Diminution of K. The second being the usual and typical result, the first and third being less frequently observed. Prolonged tetanisation (five minutes) modifies the A and K currents in a similar direction, causing a diminution (but sometimes an aug- mentation) of the A current and an augmentation (nearly always) of the K current (2424). Fria. 11.—Effect of tetanisation on A and K (2424). Thus it will be seen that the two groups of results, although not absolutely coincident, are in reasonable agreement, the two points of difference being that an augmentation of the A current has been more frequent by tetanisation than by CO,; while a diminution of the K current, rarely observed in consequence of the full action of CO., has been still more rare (once only, and that not very markedly, 2287) in consequence of tetanisation. Of these several effects the most characteristic has been the augmentation of K (2424, fig. 11) with a consequent diminution of the quotient A/K. And although—in correspondence with the not infrequent augmentation of A, there has been not infrequently an augmentation of A/K—this latter augmentation has generally been slight or even doubtful as compared with its opposite. I have been led to admit diminution of A/K as typical (2424, fig. 11, 2425 2427), and a distinct augmentation of A/K as exceptional (2387, 2388, 2393) or doubtful. (A similar augmentation of A/K by pre- dominant augmentation of A has not hitherto come under my observation in consequence of the action of CO,.) upon the EHlectrotonie Currents of Medullated Nerve. 87 Fie. 13.—Effect of CO; on A (primary diminution, secondary augmentation) (2200). | Sa are a EN Fic. 14.—Effect of tetanisation on A (augmentation) (2295). t 88 Dr. A. D. Waller. Influence of Acids and Alkalis Fra. 15.—Effect of tetanisation on A (diminution) (2296). The chief results of these experiments (and of those on tempera- ture*) are to the following effect :— The katelectrotonic current is augmented in consequence of (a) rise of temperature, (b) acidification, (c) tetanisation. It is diminished by basification. Its augmentation by tetanisation gradually declines during repose. The anelectrotonic current is diminished in consequence of rise of temperature. It is augmented by “very slight” acidification and by tetanisation, diminished by “slight” acidification and tetanisation. The characteristic effect of the presumably ‘‘ dissociative” in- fluence of rise of temperature, of acidification, and of tetanisation, is a diminution of the quotient A/K. A slighter and less assured effect of tetanisation consists in an augmentatiou of the quotient A/K. Note.—The foregoing observations form part of an investigation of the action of reagents on nerve, towards the expenses of which a erant was made by the Physiological Sub-committee of the British Association to Miss 8S. C. M. Sowton, acting as my assistant in the prosecution of the research. Our experiments during the last year have fallen under four heads :—(1) On the action of acids and alkalis upon action-currents; (2) on the action of acids and alkalis upon electrotonic currents; (3) on the action of temperature upon electrotonic currents; (4) on the action of anesthetics, of neutral salts, and of alkaloids upon electrotonic currents. The first and fourth of these four groups are not sufficiently advanced for publication, and have required to be prefaced by the second group which is now reported upon. The third group is briefly reported on in ‘ Roy. Soc. Proc.,’ Dec. 17, 1896. I wish to acknowledge Miss Sowton’s active participation during the past year in the work above specified. EKxperiments under headings (1) and (4) will, it is to be hoped, be sufficiently advanced for publication during the coming year. * © Roy. Soc. Proc.,’ vol. 60, p. 383. 89 %Q9 Aq V FO uOYRUoMISNE ArvUTT te Currents of Medullated Nerve. 0-GI- he Laan 0-8I- eid seems) 0-€2—- a ene 0:&2- ety eae o£ a “SULLU Z “00 %Q9 Aq YF Jo uorvyUoWsSNR LveuUII sy | 9. gL= | see [BULLO NT G61z 5 nr PT A a SS eT TT im) a3 ss 8. F 0.9 — 0-2 + SI—2ZT S G3 G8 = 0-8 + ‘su ~§—Z% RQ ss £00 Ss SS sss 2 “290 Aq quorjonb y/y oy} Jo wormuTeAt ¢.9 08 — | ¢ 61+ [BULL NT 8916 i=) ea: .. ae eee ae ee ee se. ene ce ern ae ) *SyABULOY ‘w/V vr | ‘V “OUT, ‘ON 9° (‘wo [ Jo youo soouvystq ejod-vayuy pus -109UT) ‘IL99 9yuvpoery ouo Aq poyoAoad syuaranH Y pues VW oy} UO sT[VyTy pur sproy jo oouenguy—'] 9/q¥, Influence of Acids and Alkalis Dr. A. D. Waller. 90 0-9 | 0-€ - | 0-81 + 02/N “OS*H ‘qooyo poy«vul-][] g. 2, | eG = | 0.61 + (ERRION TSE? 9-T | ee a | 0.361 + SBME TOT ‘ut T LoZ FOOD 1-3 | ch | c-6 + [BULL NT 982 EE a I EE EE RE SE es ee: 9. 0-7 — @.7L+ eB L “y/V¥ Jo wormurmrg LT B. leer a2 + ‘suIME g—Z | "UIUL T 10F “OO 0.¢ | g-L - | g.4 + [BULON C8ZS 0: €3 + * OL 0.S9¥+ mae G-6 + "SsUIUL & ‘UTUL T 1OF QD (ST Sy) 099 4q y jo uoynutmip AremITg oe | os | 0-12 + [BUA.0 NT 00Zz *syIVULOY a/v ONL | uy ‘OUILT, ‘ON 978IG “panurjuoo—] 2[Q% I, bs upon the Electrotonic Currents of Medullated Nerve. 1-9 | Sal ae | 0-0L+ T949FV 02/N “OS*H ‘qooe poyVul-[]] 0-9 | 0.3—- | 0-2E+ [CULO Ny 9cez a ES | RR TTS G-6 | a | 0-11 + o1/N TOS*H 6h | 08 -— | Git GSE ‘prow ormnydyus 4q y/y Jo uorynurEMg | ~~ — “y jo uvyy coqvoas Apoatzepor ST 02z/N “HOtN Y Jo uorynuruip oy4 ynq ‘poysturutp AToynTosqe ore Y pus y Wo ‘epos Gg. 86+ 6-% O28 = 0-&¢6 + 8-& ah = O- LT + GG: 08 = 0-81 + 8-1 2.0L g.8T+ y-1 Sel c.LT+ L:T Q:0T= O- ZT + 6-T On st v1 Ono = 0-OL + ‘suIU g OF FOOD — eouly — G.9Tt+ &-9 Ot. = 0-61 + L& O93 = G-8I+ ‘suru g tox anodva *Fy Ny “yue0 aod g 0-€ ee | ¢- 61 + 8-6 Of = G-6.+ ‘sult ¢ tof anodva * FANT “yuUeO aod T 09 - | 0-61 + 6 “STU ae €--0 [BULLO NT (9 8g) 6—82Pz Influence of Acids and Alkalis Dr. A. D. Waller. 98 Gg. Oy = 0. ¢¢ + OZ 18 [10 “HY poyuousne pue VY pojuctisny | gg 0-9 — 0-18 + : 8.8 o6 Q. Te + ‘s}IUN OZ 4B 4B TOD = ‘y/V pequeusne pure y poyuoutony 8.2 0-6 - G.cz+ 8 oh = Gg. FT + “y/y peyuewsny “M peysturarg = ‘s}iun Og 4B [10D c9.% qg¢- G.PL+ G-IL+ “V pesvosour ATWGSITG “Og 9" [10D a ai 0- OT + 0-8 + "V poysrurearp APQYSITG “Og 9" [10D w v 0-01 + j Gc. L os oe "20939 PoyIeUl ON “OG 4@ TION 24 hee "2 | 0-2 + "V poysturaniqg “og 7 [109 ~ = 0.61 + G.PE + "V peysomsny "0g 98 119) " " G. Te + a 53 “HM pequousne Ay4ysiTg “og ge Too) tt qi aes os *SYACROUOIT . SE/V I ‘Vv eee (‘WO T[ JO ove seouvysiqg rejod-vaquy pure -104UT) ee es ee 686 88&S L8&S 6666 "ON 94¥%[ qd ‘11299 eyourpoory ouo Aq poyoaord squering y pue Vy sy} UO TOTVESIURIOT, SOINUITY OAT JO 0.06 + eno yee SCrs 8.9 ).¢.—— ™ G.PI+ chips Nose 0B 9 [109 8-1 G9 — Gl. 11+ a0 (11 “3y) ‘qooypo yeord Ay, “yy/y poysturmrgy “yy poyuousny “y poysiuiicy G. ST 0-L - G¢. ST + O.LOJOT PCPS v-P Oo. & = 0-66 + 1045 V OS I TOD “YH pus y poywomsny | Z.F qb — 0-61 + O10 fo S2PS v1 0- ES gucT + qOoTY 0S TOD “M/¥ poysturunqg =“ peywowsny LT 0: Gs = 0-ST + SMO} OED G6ES VG ¢-9.— G. ST + TOF "OZ 1 10D = “Joo fo poy retu-TIT aad c: 0. Capiler OEE f FESS 9. § Oh = GPT + 194FV 0% 1% [10g “Y/Y poyuousne pur p poquomsny 0-8 0-7 - 0. 3L+ oLO FO 68S 0-€ ¢: 01 0-26 + L9qFV 06.98 10D M/V poysturmrg = “yy pur py peyuowmsny 6:6 Ble 0. 66 + bie x «I 1686 L-& Oe cOLs aE ; ‘0S [OQ “YW pus y poywousny | 2.¢ a. = 0. ST + etojog =| 068% VOL. LXII. 100 Mr. W. Gardiner. The Histology of ithe Cell Wall, “The Histology of the Cell Wall, with Special Reference to the Mode of Connexion of Cells. Preliminary Communica- tion.” By WALTER GARDINER, M.A., F.R.S., Fellow and Bursar of Clare College, Cambridge. Received August 11, 130%, : Since 1883* I have repeatedly endeavoured to discover some refined and generally applicable method by means of which the fine fibrille, or “‘connecting threads,’ traversing the cell membrane might be identified with certainty, and the fact of their existence settled beyond dispute. I was also anxious to be in a position to investigate the development of the threads in endosperm tissue. My researches met with little encouragement until 1894, when I succeeded in find- ing a new method, by means of which I obtained excellent results with the young developing endosperm tissue of Tamus communis. This I have further elaborated, so that either the original method, or modifications of it, can be applied to tissues generally. In the present communication I propose to give a brief account of my researches, leaving a more detailed description to a future occasion. The methods used by earlier observers for the investigation of connecting threads are essentially based upon those of Sachs and of Hanstein, by means of which they demonstrated the characteristic structure of sieve-tubes. Tangl’s important results of 1880, which ' were confirmed and extended by myself in 1883, were in fact obtained by Hanstein’s method as such, and are the outcome of quite special conditions, and a happy combination of circumstances, depending upon the fact that in dry ripe seeds the tissue is so poor in water that with the iodine and Schulze’s solution (Chlor. Zine. Iod.), the cellulose fails to give the usual blue colour, and thus allows the darkly stained threads to come into view. The method ceases to work with unripe seeds, or even with ripe seeds which contain a certain percentage of water, and with ordinary tissue is quite useless. Certain modifications of the method of Sachs, and the method of Hanstein, which may be described in general terms as involving both a more regulated application of the swelling agents and the use of aniline dyes in place of iodine, were first and independently intro- duced by Russow and myself in the years 1882 and 1883. As regards my own researches, my first results in 1882 were obtained by a modi- fication of Sachs’ method, and consisted in swelling sections of fresh tissue with sulphuric acid, and then staining with Hofmann’s violet (methyl violet) washed out with glycerine, or with Hofmann’s blue dissolved in pieric acid (picric Hofmann’s blue). This was succeeded * Gardiner, ‘ Roy. Soc. Proc.,’ No, 229, 1888. with special Reference to the Mode of Connexion of Cells. 101 in 1883 by a second method, which was a modification of that of Hanstein, and consisted in treating sections of fresh material with iodine solution, swelling with Chlor. Zinc. Iod. and staining with picric Hofmann’s blue or methyl violet. Certain results obtained by this latter treatment were so promising that in my final paper in 1883, with the customary rashness of youth, I described the method as being ‘‘ perfectly satisfactory ;” but no long period elapsed before I found in practice that it was of but limited application. Speaking generally, and excepting Poirault’s* researches, I think one is justified in saying that since 1883 little or no advance has been made in the improvement of methods, and that later observations rest mainly on small modifications of the methods of Russow and myself. The careful and detailed work of Kienitz-Gerloff+ unfortunately serves in great measure to demonstrate the unsatisfactory nature of the results obtained by the sulphuric acid method, and to prove that unless the threads are of exceptional size, as in Viscum album, or as in sieve-tubes, the method is unreliable. The above remarks equally apply to such of -my own results as depend upon the sulphuric acid modification. An advance was, however, made by Poirault. In place of experi- menting on sections of fresh tissue, he killed and hardened pieces of tissue in dilute iodine solution, and from the preserved material he then cut sections, which were swollen with Chlor. Zinc. Iod., or sulphuric acid, and stained either with eosin, Poirrier’s acid brown, methyl violet, crocein, or aniline green. Poirault’s researches are limited to the ferns and other vascular cryptogams, and le buried, so to speak, in his paper “‘ Anatomical Researehes on the Vascular Cryptogams.” While certain of his figures are, perhaps, not entirely convincing, the results of his research are most important, and of great interest. J am ashamed to say that I was unaware of the existence of this paper until the autumn of 1895, which was a year after I had elaborated the main lines of my own method, and applied it with success to the study of young endosperms. The great merit of Poirault’s modification is that here for the first time provisions are made for preserving and hardening the tissue before taking sections. New dyes are also used. With certain kinds of tissue this method appears to have given excellent results. _ I may now introduce my own researches. In the course of obser- vations on this particular branch of cytology, certain salient facts come to the fore. In the first place one learns that material pre- served in alcohol does not appear to be suitable for the investigation, and consequently fresh tissué has been used. Secondly, that it * Poirault, ‘ Ann. Sci. Nat. (Bot.),’ vol. 18, 1893. t+ HKienitz-Gerloff, ‘ Bot. Zeit.,’ 1891. 12 102. Mr. W. Gardiner. The Histology of the Cell Wall, appears generally necessary to bring about a definite swelling of the cell wall. Thirdly, that it is not easy to stain and isolate the threads, even when they are known to be present. These facts place many obstacles in the way of successful research. The difficulties attending the manipulation of fresh tissue are sufficiently obvious, and are apt to be so increased by the subsequent swelling as to render any really refined investigation well-nigh impossible, and as long as the threads cannot be stained so as to stand out clearly from the rest of the wall their identification is out of the question. These difficulties, which are sufficiently pronounced in the case of peculiarly favourable material such as that of endo- sperm, are only magnified when the investigation is concerned with ordinary or young tissue. In addition to the drawbacks already mentioned, the existing methods of research hitherto in use make ro provision for the preservation of tissue. It became obvious, therefore, that if the inquiry into the relations of the cell wall and the connecting threads was to be prosecuted with success, a more refined method must be devised, which conld be reduced to terms of the usual procedure, viz., killing, fixing, hardening, preserving, cutting section, staining, and mounting, and _ that the methods heretofore in use were too coarse for so delicate an investigation. I do not propose in the present paper either to give an account of the discovery of my method, or to go into elaborate technical details. It is sufficient to say that, expressed in the simplest terms, the method appears to depend upon the use of two prin- cipal reagents, viz., the osmic-acid-uranium-nitrate mixture of Kolossow as a fixative, and safranin as a dye. As a preservative I have used thymol water, and have obtained excellent results with material which has remained in it for as long a period as three years. Sections may be cut by hand or with the freezing microtome. The fixing and staining reagents must be introduced and employed in different ways, the exact manner of procedure depending upon the character and age of the particular tissue under observation. This can be best illustrated by means of definite examples, and since the whole method is somewhat complicated, it will be expedient to consider under separate heads (1) the killing and fixing, with which is also associated the swelling, and (2) the staining. In material such as that of young endosperms (e.g., the endosperm’ of Tamus communis), no swelling is required, and the tissue, cut into small pieces, may be both killed and fixed at one and the same time by Kolossow’s reagent, and then preserved in thymol water for future use. Where cnly slight swelling is necessary, treatment with water may precede that of Kolossow’s reagent. In certain classes of tissue, where the walls are swollen with comparative ease, such as with special Reference to the Mode of Connexion of Cells. 103 that of the ordinary vegetative tissue of Phaseolus vulgaris, Tamus communis, Nerium oleander, Salisburia adiantifolia, &c., small pieces are killed and swollen in an aqueous solution of picric acid, and then fixed in the Kolossow’s reagent and preserved in thymol water. Finally, where the tissues are more resistant, as, for instance, in Robinia pseudacacia, Prunus lawrocerasus, or Aucuba japonica, treat- ment with picric acid may be followed by more severe swelling by means of solutions of zinc chloride or sulphuric acid, to be succeeded as before by fixing, hardening, and preserving. The blackening of the cell contents caused by osmic acid may be removed by bleaching. From such preserved material sections may be cut when required. The process of staining is no less complicated than that of killing and fixing, und is best considered under two heads, viz.:—(1) The methods applicable to certain endosperms and tissues of similar character. (2) The methods applicable to the majority of tissues. In certain special cases it is possible to stain the threads directly either with safranin alone or by introducing safranin by means of a somewhat intricate substitution method such as that which I used with excellent results in the case of the endosperm of Tamus com- munis, where the sequence of staining was Hofmann’s blue (or soluble water blue), methylene blue, safranin, and in which more- - over the Hofmann’s blue was dissolved in dilute picric acid or uranium nitrate, and the methylene blue in dilute salt solution. Once stained with safranin, all sorts of modifications are possible. Thus, the safranin may be succeeded by gentian violet or by eosin, and with gentian violet Gram’s method is applicable and most advan- tageous. As safranin forms a precipitate with chromic acid, sections stained with safranin may be treated with this reagent, and then with silver nitrate, thus effecting a silver staining of the threads. Silver nitrate itself also forms a precipitate with safranin. In a.. cases the staining is practically limited to the threads. When the above methods of direct selective staining are applie| to ordinary tissue they are found to fail, fer it usually happens that the whole of the wall becomes deeply stained, so that the threads . are no longer visible. I was for some time completely baffled by this circumstance, but I ultimately adopted the well-known method of staining and washing out, using for the purpose orange G or acid fuchsin. With ordinary tissue the staining appears to be more easily ‘accomplished than with the thick mucilaginous walls of endosperm cells, and the method may be somewhat more syncopated. Hxcellent results may be obtained by staining at once with safranin and wash- ing out with orange G. This may be followed by staining with gentian violet, succeeded by treatment with acid fuchsin, or the sequence of staining may be safranin, gentian violet, acid fuchsin. Substitutions in which safranin, gentian violet, and eosin are included 104. Mr. W. Gardiner. The Histology of the Cell Wall, give good results. The method of staining indirectly by washing out may also be applied to endosperm tissues generally. The stained sections are best examined either in water or im very dilute glyce- rine. I have as yet given little attention to the question of making permanent preparations, although I have initiated certain experi- ments which may possibly lead to a satisfactory result. Itis probable that the method in its present form will not be found to be available for the study of adult lignified or suberised cells, though up to the present I have made no observations upon such tissues. The investigation of young tissue will, however, doubtless give good results, and will establish that in them, also, the general structure prevails. I am strongly convinced that the above method, or stmilar methods, based upon “non-alcoholic” treatment, will be found to be peculiarly suitable for the investigation of the tissues of plants, and will in the future lead to observations of interest and importance. A summary of results may now follow. As my new method owes its origin to a study of the cells of endosperm I propose to deal with that tissue first. The present investigation entirely confirms and extends the results I obtained in 1883.* At that time, however, many important problems still awaited solution. In the first place, what was believed to be the typical and universal structure could only be demonstrated with particularly favourable material and in a limited number of plants. Further, even in ripe seeds, where treatment with iodine proved that connecting threads were present, special difficulties were experienced when attempts were made to stain them with aniline dyes. Lastly, for young seeds the methods were quite unsuitable. With the present methods we are in a position to investigate the structure of endosperms generally, and to follow their development even from the earliest stages. The more refined method also gives more sure and satisfactory results. It is possible to make certain general statements concerning the connecting threads of endosperm cells. In the first place, the his- tological structure of endosperm establishes a point of great im- portance which is only emphasised by the study of tissues generally, viz., that in pitted cells the pit-closing membrane is invariably traversed by threads. For descriptive purposes these may be called “‘pit-threads.” Threads may also be present which traverse tie general wall, and these may, similarly, be called “ wall-threads.” In the somewhat exceptional cases of unpitted cells the thread system is necessarily composed of ‘‘wall-threads” only. In many pitted cells both “ pit-threads”’ and “ wall-threads ” are present; but im the majority of cells the threads appear to be limited to the pits, * Gardiner, ‘ Roy. Soc. Proc., No. 225, 1883. with special Reference to the Mode of Connexion of Cells. 105 though it is not improbable that even in such cases other threads traversing the general wall will be found to occur. Where both “ pit-threads”’ and “ wall-threads’’ coexist in one and the same cell, the former are stouter and more readily stainable than the latter. In pitted cells the pit-threads are necessarily in groups, and it is a point of some interest that the wall-threads also are usually in groups—as though a pit were present. This is especially striking in such cases as the unpitted cells of Tamus communis (fig. L) and Hordeum vulgare. In Tumus communis, while the side walls exhibit the usual arrangement of isolated groups of threads, the end walls are traversed by a single large group, as in sieve-tubes. The structure of the endosperm cells of Inlium martagon (fig. 2) is Ul ( Ura ATT et i fcc it a Ay 5 os =u of some interest. The cells are pitted and each pit has its group of “pit-threads.” In any given group the threads are arranged in bundles, recalling the similar arrangement of achromatin fibres, which Strasburger and others have found to accompany nuclear divisions in this plant. The mode of development of the threads 106 Mr. W. Gardiner. The Histology of the Cell Wail, and the phenomena which accompany germination were iuvesti- gated in the endosperm of Tamus communis only. The threads are found to be present at a very early stage, and can be detected even in the very youngest and thinnest walls. They are at first uniformly distributed over the cell membrane. In the case of the side walls, as surface growth proceeds, small groups of threads become separated from one another by intervening areas of clear membrane, while in the end walls, where the extension of surface is less, this segregation into groups does not take place. In the early stages the growth in thickness of the wall is not uniform, and pits are formed on the side walls at those points where thread groups occur; but they have cnly a transient existence, and ultimately disappear. It is, however, interesting to note that the vegetative tissue of Tamus commuits consists for the most part of pitted cells. In, germination the ferment, in the first instance, appears to be conducted into the wall by means of certain of the threads, but when once an entrance is effected the corrosive action rapidly spreads quite independently of the threads, becoming the more potent as it reaches the neighbourhood of mucilaginous and less resistent middle lamella. In a given wall the penetration commences simultaneously at several centres, and at each centre the affected areas assume the general form cf small cones with their apices directed towards the cell lumen (fig. 3). Moreover, since the action of the ferment soon extends from cell to cell in two adjoining cells, where the common wall is affected on both sides, each side havimg its cone, the base of one cone appears opposed to that of the other. At this stage, by appropriate staining, the threads may still be seen shining through the disorganised mucilage of the affected areas. As the ferment action proceeds the boundaries of the several areas con- tinually extend and at length unite when the whole of the wall is involved. The disorganisation of the wall is accompanied by marked stratification. The sphere of infiuence of the ferment action with special Reference to the Mode of Connexion of Cells. 107 is curiously limited and hardly extends beyond the immediate neigh- bourhood of the absorbent foot of the young embryo. The mode of disappearance of the wall by the coarse corrosive action of the ferment, and the manner of its proceeding with so little relation to the threads, seems to indicate that the structure of endosperm cells has more relation to the conduction of impulses and of food supply than to the needs of germination. Satisfactory as are the results derived from the study of endo- sperm cells, the real success was achieved when the method was so modified that it could be applied with certainty and success to the investigation of ordinary tissue, and to specialised forms of it, such as pulvini and tendrils. The unsatisfactory character of the somewhat meagre results hitherto obtainable have not, I think, succeeded in generally estab- lishing a conviction that the structures postulated have necessarily an actual existence. Botanists have, on the contrary, been inclined to look askance at results depending upon such statements as ‘a stained area” or ‘“‘a doubtful striation,” and there lurked in the minds of many, the suspicion that the histology of endosperms was possibly exceptional, and peculiar to.that tissue. For the future such doubts need no longer be entertained, since it is now easy to demonstrate that the structure exhibited by endosperm tissue is in all respects entirely typical of plant tissue generally. In these days of active investigation, it is not often given to one to be the medium for the criticism of his own research. This good fortune, however, now falls to me, and I hasten to say that in the light of the present investigation it is quite clear that apart from endosperms, and with such exceptions as Aucuba japonica, my earlier work on the continuity of the protoplasm, in pulvini and other tissues, does not affurd absolute proof that a communication between cells actually occurs, but for the most part only brings forward strong evidence that such connection is exceedinyly probable. At this juncture, also, I note, with satisfaction, that the results then obtained were able to save me from the error of a belief in the existence of a system of open pits which have since then been repeatedly figured and desciibed, and against which I have persist- ently spoken. As I have already remarked, my new method makes it possible to establish with certainty that the structure of all kinds of plant tissue is precisely similar to that of endosperms, and that exactly the same modifications are exhibited. In pitted cells every pit- closing-membrane is traversed by its group of threads, and in un- pitted cells similar groups also occur. In square-ended tubular cells, such as those of the leaf stalk of Lilium martagon (fig. 4), the 108 Mr. W. Gardiner. The Histology of the Cell Wall, Fia. 4. numerous small isolated groups of wall-threads are present on the side walls, and a large group occupies each end wall. In certain cells both “pit-threads” and “wall-threads’” may simultaneously be present. The method gives equally good results with thin or thick- walled tissue, and in the case of the thinnest walls an “en face” view can be obtained where a sectional view fails. ‘The threads vary in size. They are, for instance, exceedingly thick in Viscum album, where they are seen with the greatest ease; they are well developed in Phaseolus multiflorus and Liliwm martagon, and they are fine and delicate in Aucuba japonica (fig. 5). The connecting threads may be Fia. 5. observed equally in the epidermis and cortex, and in all the living cells of the tissues of the central cylinder. In a section of any given tissue, it can be readily seen that all the cells are placed in commu- nication with one another. One is tempted to expatiate upon the wonderful beauty of the appearance presented by the walls studded with their pits, each pit with its closing membrane traversed by the thread complex after the manner of sieve-tubes. It is indeed a sight which cannot fail to convey a lasting impression of wonder and surprise. The pit-threads of such pulvini as were examined presented no striking difference, either in appearance or distribution, to the similar structures in ordinary tissue, and the same appears to be true of the sieves of tendrils. The important point, however, is this, with special Reference to the Mode of Connexion of Cells. 109 that in all these tissues the threads perforating the pit-closing mem- brane can be readily seen, and even counted. A point of interest was observed in the epidermal cells of Tamus communis and Lilium martagon (fig. 6), viz., that the external or free walls are penetrated by a system of threads which radiate from the cell lumen and extend to the cuticle, so that the latter is the only structure intercalated between the protoplasm and the environment, The important bearings of this observation are obvious. Fie. 6. The following is a list of the tissues which have up to the present been investigated, together with the date of observation. In each case connecting threads were observed. In 1894. The young and developing endosperm of Tamus communis and the young endosperm of Asperula odorata.. In 1895. The cotyledons of Tropeolum majus. The endosperm of Lilium martagon and Fritillaria imperialis. The root of Ranunculus astaticus. The leaf stalks of Tamus communis, Viscum album, and Marattia elegans. In 1897. The leaf stalks of Aucuba japonica, Prunus lawrocerasus, Nerium oleander, Lilium martagon, Lilium candidum, Salisburia adiantifolia, and Asplenium rutefoiium (fig. 7). The flower stalk of \ Big. 7. 110 Mr. W. Gardiner. The Histology of the Cell Wall, Taraxacum dens-leonis. The perianth-leaves of Liliwm martagon. The pulvini of Mimosa sensitiva and Robinia pseudacacia (fig. 8). The tendril of Cucurbita pepo. The endosperm of Hordeum vulgare. Fie. 8, Although the research is still in its preliminary stages, since my attention has been chiefly directed to the elaboration of the new method, yet the results already obtained are sufficiently numerous and suggestive to enable one to make certain remarks and general statements. It is impossible to resist the conclusion that the connecting threads consist either wholly or partly of protoplasm, and this view is largely confirmed by the staining reactions. It is, however, not improbable that the protoplasmic filament may be sievontded by a mucilaginous sheath. Osmic acid induces no blackening as it does in the threads of many sieve-tubes, and I am inclined to believe that in ordinary tissue the threads consist for the most part of ectoplasm, and are to be regarded in fact as extensions of it. The threads appear to be present ab witio. This fact, coupled with the surface-growth of the cell wall, furnishes a sufficient expla- nation of the “barrel figures” so generally assumed by the various thread groups. The resemblance to the similar figures which accompany nuclear division is therefore superticial. Nevertheless it seems certain that the threads do, as a matter of fact, arise from that part of the cytoplasm which at the period of formation of the cell plate forms the fibres of the so-called “ nuclear spindle,” and that these fibres become, so to speak, partiaily imprisoned in the young wall. My results appear to indicate that in a given cell the whole system of connecting threads arise at this early stage, and that no subsequent development occurs. This statement will, however, require careful confirmation, and has certain bearings on such iateresting questions as the theory of grafts and “sliding growth.” In dead cells, such as those of ripe endosperm, the threads appear to degenerate into mucilage, and this is possibly also the case 1 in adult lignified and other cells. In the particular tissues which I have inveshieiems the threads can be shown to be present in all cells which still retain their cellu- luse character, and although I have not actually succeeded in J with special Reference to the Mode of Connexion of Cells. 111 observing them in adult lgnified and suberised tissue, it seems certain that they will be found there also. There can be little doubt that they occur universally in the cells of all the tissues of all plants. From this arises the fundamental conception that the plant body must be regarded as a connected whole, and that the cell walls occupy only a subservient position. Thus our views as to the ulti- mate histology of tissue must be considerably modified. A new vista also opens to cytological research in the direction of the accurate determination of the distribution and orientation of the threads in the yarious tissues, which can hardly fail to lead to important results. Should the structure presented by the external walls of the epi- dermal cells of Tamus communis and Liliwm martagon be found to be of general occurrence, we shall be prepared for most interesting results when the examination is extended to secreting gland cells: to such non-cellular organisms as certain alg and fungi, and to such unicellular bodies as spores and pollen grains. Two important functions are, doubtless, performed by the con- necting threads, viz., the conduction of impulses and the conduc- tion of food. As to the first, there can be no question; and as to the second, one cannot but reflect that it must be of the ereatest advantage to the plant to be able to transmit from cell to cell as occasion requires, and in a definite and determinate direction, highly organised food supplies and even protoplasm itself. It is, of course, possible that in the threads themselves a definite division of labour may occur as regards the transmission of food and the con- duction of stimuli. The consequences which arise from our more perfect knowledge of plaut histology are obvious and far reaching. In the first place we learn that the structure exhibited by sieve- tubes, which in the past was regarded as peculiar, is shown to be typical of cells generally, with this slight difference, however, _ that in sieve-tubes a secondary eulargement of the pores appears to occur. All the cells of a tissue must be regarded as being connected together by delicate groups of protoplasmic threads, which traverse either the general wall or the pit-closing membrane, just as the sieve-tube threads traverse the so-called ‘“‘sieve plate,” a fact which at length enables us to do tardy justice to the dominant position occupied by the protoplasm of the plant body, and to understand how the deep-seated cells of a tissue can telegraph their needs to those at the periphery, cell after cell taking note of the wording of the message, or how the peripheral cells may communicate to the interior their sense of gravity, light, heat, or touch, to which the whole organ may reply as its peculiar organisation directs. As an integral part of cell structure, the connecting threads con- 112 Lord Rayleigh. stitute a factor, which cannot fail to have an important bearing in all general questions, such as the growth of the cell wall, the conduction of food, the ascent of water, the process of fertilisation,. the penetration of fungi into their host, the process of secretion, and the transmission of the impulses which determine growth and move- ment of plant organs. Concerning certain of these problems, I should like to make a few concluding remarks. As to the passage of water from the root hair to the vessel, the presence of connecting threads in the cells of root tissue makes it possible to imagine that the ordinary laws of osmosis may be pro- foundly modified, and that the filaments which establish protoplasmic — continuity may conduct stimuli, leading, for instance, to a difference in reaction of the proximal and distal halves of any given cell. Similarly, it is conceivable that a definite polarity is established, which helps to determine the direction of the flow. As to the larger question of rapid water movement, althongh this is neither the time nor the place to enter into theory, yet I cannot refrain from remark- ing that it is not impossible that the threads, doubtless present in large quantities in dead vessels, may, if they suffer mucilaginuus change, have some bearing on the question, e.g., by assisting to sustain the water at any given level or attracting water in the imme- diate environment. In any case, I am strongly of opinion that the part played by mucilage and the force of hydration have not as yet received sufficient attention. As to movements generally, I am still unable to accept Pfeffer’s view of the subsidiary part played by the protoplasm in connexion with turgidity,* and I am still of opinion that the ectoplasm is the master factor which determines the condition of the cell. The present research demonstrates among other things that there are fixed points in the ectoplasm, and this may have some bearing on the possibility of estab:ishing the periodic or sudden contractions and dilatations which I believe are associated with turgescence, and of which such a phenomenon as the effusion of water from the cells of a stimulated Mimosa pulvinus is but an abnormal instance. “On the Viscosity of Hydrogen as affected by Moisture.” By Lorp RAYLEIGH, F.R.S. Received September 8, 1897. In Sir W. Crookes’s important work upon the viscosity of gasest the case of hydrogen was found to present pesuliar difficulty. ‘With each improvement in purification and dryinz I have obtained * Gardiner, ‘ Ruy. Soc. Proc.,’ vol, 43, 1887. ¢ ‘Phil. Trans.,’ 1881, p. 387. ° On the Viscosity of Hydrogen as affected by Moisture. 113 a lower value for hydrogen, and have consequently diminished the number expressing the ratio of the viscosity of hydrogen to that of air. In 18761 found the ratio to be 0°508. In 1877 I reduced this ratio to 0°462. Last year, with improved apparatus, I obtained the ratio 0°458, and 1 have now got it as low as 0°4439”’ (p. 425). The difficulty was attributed to moisture. Thus (p. 422): “ After work- ing at the subject for more than a year, it was discovered that the discrepancy arose from a trace of water obstinately held by the hydrogen—an impurity which behaved as I explain farther on in the case of air and water vapour.” When occupied in 1888 with the density of hydrogen, I thought that viscosity might serve as a useful test of purity, and I set up an apparatus somewhat on the lines of Sir W. Crookes. A light mirror, 18 mm. in diameter, was hung by a fine fibre (of quartz I believe) about 60 cm. long. A small attached magnet gave the means of starting the vibrations whose subsidence was to be observed. The viscosity chamber was of glass, and carried tubes sealed to it above and below. The window, through which the light passed to and fro, was of thick plate glass cemented to a ground face. This arrange- ment has great optical advantages, and though unsuitable for experi- ments involving high exhaustions, appeared to be satisfactory for the purpose in hand, viz., the comparison of various samples of hydrogen at atmospheric pressure. The Toppler pump, as well as the gas generating apparatus and purifying tubes, were connected by seal- ing. But I was not able to establish any sensible differences among the various samples of hydrogen experimented upon at that time. In view of the importance of the question, I have lately resumed these experiments. If hydrogen, carefully prepared and desiccated in the ordinary way, is liable to possess a viscosity of 10 per cent. in excess, a similar uncertainty in less degree may affect the density. I must confess that I was sceptical as to the large effect attributed to water vapour in gas which had passed over phosphoric anhydride. Sir W. Crookes himself described an experiment (p. 428) trom which it appeared that a residue of water vapour in his apparatus indicated the viscosity due to hydrogen, and, without deciding between them, he offered two alternative explanations. Hither tle viscosity of water vapour is really the same as that of hydrogen, or under the action of the falling mercury in the Sprengel pemp decomposition occurred with absorption of oxygen, so that the residual gas was actually hydro- gen. It does not appear that the latter explanation can be accepted, at any rate as regards the earlier stages of the exhaustion, when a rapid current of aqueous vapour must set in the direction of the pump; but if we adopt the former, how comes it that small traces of water vapour have so much effect upon the viscosity of hydrogen P 114 Lord Rayleigh. It is a fact, as was found many years ago by Kundt and Warbure* (and as I have confirmed), that the viscosity of aqueous vapour is but little greater than that of hydrogen. The numbers (relatively to air) given vy them are 0°5256 and 0°488. It is difficult to believe that small traces cf a foreign gas having a 6 per cent. greatcr viscosity could produce an effect reaching to 10 per cent. In the recent experiments the hydrogen was prepared from amalgamated zinc and sulphuric acid in a closed generator con- stituting in fact a Smee cell, and it could be liberated at any desired rate by closing the circuit externally through a wire resist- ance. The generating vessel was so arranged as to admit of exhaustion, and the materials did not need to be removed during the whole course of the experiments. The gas entered the vis- cosity chamber from below, and could be made to pass out above through the upper tube (which served also to contain the fibre) into the pump head of the Toéppler. By suitable taps the viscosity chamber could. be isolated, when observations were to be com- menced. The vibrations were started by a kind of galvanometer coil in connection (through a key) with a Leclanché cell. As a sample set of observations the following relating to hydrogen at atmo- spheric pressure and at 58° F., which had been purified by passage over fragments of suiphur and solid soda (without phosphoric anhydride), may be given :— Observations on June 7, 1897. cs 65 °4 ask Be = 423 °7 88°9 308 °3 2 °554 — 401 °3 110°0 312 °4: 2 °495 0°059 381-5 128-9 271 °5 2 °434 0 O61 364 °4 144-1 2359 2 372 0-062 349 °7 158 °6 205 6 2°313 0-059 336 °8 169 °8 178 °2 2°251 0-062 325 °7 180°6 155°) 2°193 0 °058 315°7 189°8 135°1 2-131 0°062 307 °2 197-3 117 °4 2-070 0°061 300 °0 204° 6 102 °2 2 009 0-061 293 °7 210°6 89 °1 1-980 0-059 287 °8 = V7? 2 1°888 0 °062 Mean log. dec. = 0°0604. The two first columns contain the actually observed elongations upon the two sides. They require no correction, since the scale was bent to a circular arc centred at the mirror. The third column * ‘Pogg. Ann.,’ 1875, vol. 155, p. 547. On the Viscosity of Hydrogen as affected by Moisture. 115 gives the actual arcs of vibration, the fourth their (common) loga- rithms, and the fifth the differences of these, which should be con- stant. The mean logarithmic decrement can be obtained from the first and last arcs only, but the intermediate values are useful as a check. The time of (complete) vibration was determined occasionally. It was constant, whether hydrogen or air occupied the chamber, at 26°2 seconds. The observations extended themselves over two months, and it would be tedious to give the results in any detail. One of the points to which I attached importance was a comparison between _ hydrogen as it issued from the generator without any desiccation whatever and hydrogen carefully dried by passage through a long tube packed with phosphoric anhydride. The difference proved itself to be comparatively trifling. For the wet hydrogen there were obtained on May 10, 11, such log. dees. as 0°0594, 0°0590, 0°0591, or as a mean 0:0592. The dried hydrogen, on the other hand, gave 0°0588, 0°0586, 0:0584, 0°0590 on various repetitions with renewed supplies of gas, or as a mean 0:0587, about 1 per cent. smaller than for the wet hydrogen. It appeared that the dry hydrogen might stand for several days in the viscosity chamber without alteration of logarithmic decrement. It should be mentioned that the apparatus was set up underground, and that the changes of temperature were usually small enough to be disregarded. In the next experiments the phosphoric tube was replaced by others containing sulphur (with the view of removing mercury vapour) and solid soda. Numbers were obtained on different days such as 0°0591, 0°0586, 0°0588, 0:0587, mean 0:0588, showing that the desiccation by soda was practically as efficient as that by phos- phoric anhydride. At this stage the apparatus was rearranged. As shown by observations upon air (at 10 cm. residual pressure), the logarithmic decrements were increased, probably owing to a slight displacement of the mirror relatively to the containing walls of the chamber. The sulphur and soda tubes were retained, but with the addition of one of hard glass containing turnings of magnesium. Before the mag- nesium was heated the mean number for hydrogen (always at atmo- spheric pressure) was 0°0600. The heating of the magnesium to redness, which it was supposed might remove residual water, had the effect of increasing the viscosity of the gas, especially at first.* After a few operations the logarithmic decrement from gas which had passed over the hot magnesium seemed to settle itself at 0-0606. When the magnesium was allowed to remain cold, fresh fillings gave again 0°0602, 0:0601, 0:0598, mean 00600. Dried air at 10 cm. * The glass was somewhat attacked, and it is supposed that silicon compounds may have contaminated the hydrogen. VOL. LXII. K 116 On the Viscosity of Hydrogen as affected by Moisture. residual pressure gave 0°01114, 0°01122, 0°01118, 0°01126, 0:01120, mean 0°01120. . In the next experiments a phosphoric tube was added about 60 cm. long and closely packed with fresh materia]. The viscosity appeared to be slightly increased, but hardly more than would be accounted for by an accidental rise of temperature. ‘The mean uncorrected number may be taken as 0°0603. ) The evidence from these experiments tends to show that residual moisture is without appreciable influence upon the viscosity of hydrogen ; so much so that, were there no other evidence, this con- clusion would appear to me to be sufficiently established. It remains barely possible that the best desiccation to which I could attain was still inadequate, and that absolutely dry hydrogen would exhibit a less viscosity. It must be admitted that an apparatus containing cemented joints and greased stop-cocks is in some respects at a disadvantage. Moreover, it should be noticed that the ratio 0:0600 : 0°1120, viz. 0°536, for the viscosities of hydrogen and air is decidedly higher than that (0°500) deduced by Sir G. Stokes from Crookes’s observations. According to the theory of the former, a fair comparison may be made by taking, as above, the logarithmic decrements for hydrogen at atmospheric pressure, and for air at a pressure of 10 cm. of mercury. 1 may mention that moderate rare- factions, down say to a residual pressure of 5 cm., had no influence on the logarithmic decrement observed with hydrogen. I am not able to explain the discrepancy in the ratios thus exhibited. A viscous quality in the suspension, leading to a sub- sidence of vibrations independent of the gaseous atmosphere, would tend to diminish the apparent differences between various kinds of gas, but I can hardly regard this cause as operative in my experi- ments. For actual comparisons of widely differing viscosities I should prefer an apparatus designed on Maxwell’s principle, in which the gas subjected to shearing should form a comparatively thin layer bounded on one side by a moving plane and on the other by a fixed plane. Electromotive Force of different Forms of the Clark Cell. 117 «On the Variation of the Electromotive Force of different Forms of the Clark Standard Cell with Temperature and with Strength of Solution.” By H. L. CALLENDAR, M.A., F.R.S., McDonald Professor of Physics, and H. T. BARNES, M.A.Sc.,° Demonstrator of Physics, McGill University, Montreal. Received August 12, 1897. § 1. Objects of the Investigation. The primary object of the present series of experiments was that of equipping the McDonald Physics Building of McGill University with a reliable and accurate set of standard cells, and not that of forming the subject of a communication to any scientific paper. In the course of the work, however, several points have come under our notice, which we venture to think may be of interest to others engaged in any investigation requiring the employment or construc- tion of such standards. Among other points, we have devoted special attention to the accurate determination of the temperature-coefficients of various forms of Clark cell; to the construction of cells free from ‘“ diffu- sion-lag ”’ consequent upon change of temperature; and to the inves- tigation of the limits of accuracy attainable with Clark cells under both constant and varying temperature conditions. We have succeeded in making a very simple modification in the Board of Trade form of Clark cell, which makes it equal to any other form in respect of freedom from diffusion-lag; and we have made several forms of cell hermetically sealed with glass and platinum, which we hope will stand the test of time better than those sealed with wax and marine glue. We have also made a special investigation of the effect of changes of strength of the solution of zinc sulphate on the E.M.F., involving determinations of the solubility of zinc sulphate and of the density of the solutions, which appear to lead to very simple formule of some theoretical interest. § 2. Preliminary Work. At the outset of our work, we were extremely fortunate in finding the laboratory already equipped by the liberality of Mr. W. C. McDonald, the donor of the Physics Building, with a very fine and complete set of resistance standards and electrical measuring in- struments, collected by Professor John Cox, under whose supervision the laboratory was planned and erected. There were in the collec- tion several portable Clark cells by Muirhead, and a set of Carhart cells by Queen and Co. were shortly added. K 2 118 Prof. H. L. Callendar and Mr. H. T. Barnes. Several comparisons of these portable cells were made early in 1894 at different dates and under ordinary laboratory conditions. The results, when corrected for temperature by means of the enclosed thermometers, showed irregular differences, often amounting to nearly 3 millivolts. Of a pair of cells enclosed in one brass case together with a thermometer, one would ke sometimes higher and sometimes lower than the other, although the utmost care was observed in using them, and they were both necessarily always under closely similar conditions as regards change of temperature. The set of Carhart cells were much better in this respect. They rarely showed irregular differences of more than half a millivolt; but one of the cells, owing to defective sealing, was generally some 2 millivolts lower than the others, and has since that time fallen still further. It appeared probable at first that these discrepancies were partly due to inequalities of temperature between the cells and their attached thermometers, and, if so, that they would be insepar- able from portable cells of such a form under these conditions. About the same time, a number of Clark cells were set up by Professor. Callendar and some of the advanced students and demon- strators, in accordance with the form described in the Board of Trade Memorandum, as figured in Glazebrook and Shaw’s ‘ Practical Physies,’ p. 576. Every detail of manipulation and construction was carefully followed, except that the cells were set up in test-tubes 6 inches long, to permit of their being more deeply immersed in water, and that the glass tubes containing the electrodes were pro- vided with mercury cups at the top to facilitate the making and changing of connections. The cells were kept immersed in water im large glass bottles provided with a stirrer and thermometer. The cells set up in this way, although made by different students. with different solutions at different dates, were found to agree more closely among themselves than the portable forms, owing probably to the more constant and certain conditions to which they were exposed. They still, however, exhibited irregular differences, even when exposed to précisely similar variations of temperature, and it was felt that they could not be used with confidence for any work in which an accuracy of one part in 10,000 was desired. Some determinations were also made of the temperature coefficients of these cells when exposed to a variation of temperature at the rate of about 10° in two hours. The results were very fairly consistent for'each cell, but gave very different values for the mean coefficient between 10° and 20° C. The lowest value obtained was 0°00045, the highest 0:00069. The value commonly given for these cells is 0:00078. They were all saturated cells containing an abundance of crystals, which remained visible on the top of the paste and through- out the mass at the highest temperature. Electromotive Force of different Forms of the Clark Cell. 119 _ Inexamining the results it was noticed that the value of the co- efficient did not depend on the quantity of crystals in the cell, but on the distance of the end of the zinc rod from the crystals. The difference of the results was, therefore, evidently due to slowness of diffusion of the state of saturation through the body of the solution. So long ago as 1886 aset of five cells was set up by one of the present writers at the Cavendish Laboratory, Cambridge, with zinc rods of different lengths, with the object of testing the effect of diffusion on the temperature-coefficient. These five cells were of the original Rayleigh pattern, with platinum wires sealed through the bottom. They could not be immersed in a water-bath, and were found to be of a somewhat unsuitable form for the experiment, which was then discontinued for more important work. Some more recent tests of this set of cells will be found in the paper on the Clark cell, by Glazebrook and Skinner.* The cells are numbered, 6, 7,8, 9, and 10 in the paper, and are among the oldest in the possession of the Cavendish Laboratory. It will be seen from the tests that they exhibit small discrepancies such as might very probably arise from differences in the time required for diffusion in the different cells. From a study of the above-mentioned paper, it appears likely, in our opinion, that the differences observed in the case of the other cells of the same type may have been affected by a similar cause. It would appear, in fact, inevitable that cells of the simple Board of Trade pattern should exhibit some effects of diffusion-lag, especially if subjected to considerable or rapid changes of tempera- ture. This form of cell, however, is so convenient to use, and so easy to make, that it seemed to us desirable to make a more careful examination of the case, with the object, if possible, of constructing cells of this form with a definite temperature-coefficient and a negli- gible diffusion-lag. In October, 1894, the class was joined by Mr. H. T. Barnes, who, as assistant to Dr. Harrington, Professor of Chemistry, had obtained considerable experience in chemical manipulation. From this date onwards the work of making and testing various forms of Clark cell has been performed almost entirely by Mr. Barnes, but the observations and calculations throughout .have been checked and verified by Professor Callendar, who has devoted special attention to the thermometry. § 3. Constant-temperature Baths. The first step in the investigation consisted in making a pair of suitable water-baths, controlled by delicate thermostats, which could * ©Phil. Trans.,’ A, vol. 183 (1892), p. 586. 120 Prof. H. L. Callendar and Mr. H. T. Barnes. be set in such a manner as to keep the temperature steady for any desired length of time at any point between 5° and 30° C. The method of regulation adopted for these baths was similar to that described by Griffiths,* but much less elaborate. The baths were made of copper, and were encased in felt and wood. They were heated by a stream of tap-water passing through a copper tube over a regulated gas flame. The regulators were made to cut off the gas so sharply that a difference of temperature of one-tenth of a degree sufficed to change the gas supply from full flame to no flame. With these very sensitive regulators, some trouble was experienced. at first, partly owing to the excessive variations of the Montreal gas pressure, and partly owing to the sudden changes of the climate. In the end, however, these difficulties were so successfully overcome that, on the longest continuous run, extending over nearly a fort- night, the temperature of the bath did not vary by so much as 002° C. throughout the whole period. The temperature of the tap-water averaged about 8° C. in mid- winter, and seldom rose above 13°C. in summer. This generally sufficed to keep the baths steadily at 15° C. evenif the temperature of the room was as high as 25°C. In order to obtain steady tempera- tures at points below 15° C., the stream of tap-water was led through a copper spiral immersed in melting snow or ice before passing over the gas flame. In this manner the baths could be set to regulate steadily at temperatures as low as 5° C. The two baths were generally set to regulate at different tempera- tures, so that by transferring a cell from one bath to the other the effect of a sudden and definite change of temperature could be ob- served. The time required by cells of different forms to reach their steady final values could thus be determined, and the effects of diffusion-lag could be readily distinguished from the immediate change of H.M.F. consequent upon a change of temperature. § 4. Electrical Apparatus. By the use of these accurately regulated water-baths, the tempera- ture of the cells became so much a matter of certainty that we found it desirable to make the comparisons to the hundredth of a millivolt, corresponding to the hundredth of a degree Centigrade of temperature. From the results of our experiments, we have reason to conclude that the Clark cell, under suitable conditions, permits the attainment of this order of accuracy, and is far superior to the silver voltameter for accurate determinations. The comparisons of the cells were made by the usual Poggendorff method, with a 6000-ohm galvanometer. The potentiometer used * © Phil. Trans.,’ A, vol. 184 (1893), p. 374. Electromotive Force of different Forms of the Clark Cell. 121 for the earlier comparisons was a long wire, having a resistance of 86 ohms, wound on a cylinder in one hundred turns. Each turn was divided into one hundred parts, and readings were taken to one tenth of a division. Hach division corresponded approximately to one five-thousandth part of a volt, a storage-cell being used to supply the steady current through the wire. This potentiometer had been accurately calibrated throughout its length at two different dates. The results agreed so closely that it could be used with confidence for measuring relatively large differences of potential with an accuracy of at least one-half division of the wire, equivalent to 0:0001 of a volt. The errors of the uncorrected wire amounted to over ten divisions in many places. For the later experiments a simpler, and in many respects more convenient, form of potentiometer was used. ‘Two resistance boxes, containing resistances adjustable up to 2000 ohms each, were con- nected by a platinum-silver bridge-wire having a resistance of 18 ohms. The wire was 2 metres long, in four lengths of 50 cm. each, with a millimetre scale, and was adjusted to read direct in volts, at the rate of 1 mm. to one-hundredth of a millivolt, in the following manner :—A resistance of 18/20 x 1420 ohms was taken out of the first box, and the resistance in the second box was adjusted to make the standard cell at 15° C. read near the point 140 cm. of the wire, 7.e., 14 millivolts above 1:420 volts. The bridge-wire could thus be used directly for measuring small diffe- rences of H.M.F. not exceeding 20 millivolts, with an accuracy of at least one part in a thousand on a difference of this order. Larger differences could be readily dealt with by transferring resistance from one box to the other in such a way as to keep the sum con- stant, each ohm transferred being reckoned at 20/18 of a millivolt. It will be understood that the resistance of the bridge-wire was carefully measured in terms of the boxes, that their temperature- coefficients were very nearly the same, and that the wire was tested for uniformity, to insure the above order of accuracy in the deter- mination of differences of E.M.F. of this magnitude. § 5. Thermometry. In working to the hundredth of a millivolt, it was necessary to xnow the temperature of the baths to the hundredth of a degree C. Two thermometers were generally used, one in each water-bath. They were both carefully compared with a platinum thermometer, and their indications were in all cases reduced to the absolute scale. One of the thermometers was by Geissler, divided to tenths of a degree. This thermometer had evidently been graduated to read temperature on the absolute scale direct. Its errors, after correcting 122 Prof. H. L. Callendar and Mr. H. T. Barnes. for rise of zero, were found to be very small and irregular, seldom exceeding 0°01° C. The second thermometer was by Hicks, divided to twentieths of a degree. Its corrections were found to be very nearly the same as those of the Kew mercurial standard. Over the range 0° to 30° C. the changes of zero of these thermo- meters would never exceed 0°01° C.,and were, therefore, disregarded. The correction for the length of stem exposed, never exceeding two or three hundredths of a degree, could be applied with sufficient accuracy when required. The comparisons were made with a platinum thermometer con- structed of special wire, which has been repeatedly tested by Pro- fessor Callendar, and also by Mr. Griffiths, and by Messrs. Heycock and Neville. The wire is the same as that used in the thermometers made for the Kew Observatory, and its “‘ delta-coefficient ” has been taken as 1°50. The resistance box used was of special design, reading to 0:0001° C. It was exhibited by Professor Callendar at the May Conversazione of the Royal Society in 1893. The box used at Kew for platinum thermometry has recently been constructed on the same model, and has been described in ‘ Nature,’ November 14, 1895. The Kew box differs chiefly in the use of plugs for mercury contacts, and in the absence of the temperature compensating coils. § 6. Comparisons of Board of Trade Cells. With this apparatus many more accurate and careful comparisons of Board of Trade cells were made. Several new cells, prepared by H. T. B. and by other students, were compared with those a year old. The newer cells were generally found to have a slightly higher - E.M.F. than the old, and in general differences of the same order as before were observed, if the cells were subjected to different treat- ment. It was noticeable, however, that B.O.T. cells, prepared about the same time in a similar manner, if kept exposed to similar stable conditions, would generally attain the same H.M.F., within one or two tenths of a millivolt, after a day or so in the constant tempera- ture bath at 15°C. The importance of keeping cells of this type ata constant temperature has been shown by Griffiths,* who has obtained very consistent results with B.O.T. cells treated in this manner. Kahle, on the other hand,+ finds differences, amounting to 4 or 5 millivolts in some cases, between the nine B.O.T. cells which he tested under constant and similar temperature conditions. Such differ- ences are quite beyond the range of our experience, and we do not * ¢Phil. Trans.,’ A, vol. 184 (1893), p. 388. tT ‘ Wied. Ann.,’ vol. 51 (1894), p. 194. Electromotive Force of different Forms of the Clark Cell. 123 think that they fairly represent the performance of B.O.T. cells under the conditions that he describes. § 7. Temperature Coefficients of B.O.T. Cells. Fresh determinations of the temperature coefficients of the old cells, in addition to those of the new cells, were made under different conditions and with greater accuracy by means of the constant temperature baths. For four of the cells in which the end of the zinc rod was at a small distance from the crystals, the mean coefficients obtained on raising the temperature from 15° to 25° C. varied from 0:00046 to 0°00051, and were in practical agreement with previous tests of the same cells under similar conditions. Precisely similar values were obtained after keeping the cells at 25° C. for the night, and then lowering the temperature to 15° C. In another case, after keeping two of the cells at 25° C. for the night, the changes of their E.M.F. from their values at 15° C. were found to be 9:0 and 8'6 millivolts, giving co-efficients 0°00063 and 000060 respectively. The greater change is evidently due to the time allowed for diffusion. _ On cooling the cells down from 15° to 0° C., allowing them to remain for an hour at the latter temperature, the mean coefficients obtained were invariably much larger. The reason is evidently that. the zinc becomes partially imbedded in the crystals at the lower temperature, and is necessarily in contact with a normal saturated solution throughout a considerable portion of its surface. In one ase a coefficient as high as 0°00075 was obtained ; in another a co- efficient as low as 0°00059. In the latter case the rod was very long, and a considerable length was probably exposed to a supersaturated solution. That this state of supersaturation is likely to occur, and to persist for a considerable time, is also illustrated by another experiment with a cell containing very few crystals. After keeping the cell in question at 25° C. for a day, it was observed that all the crystals had disappeared, whereas the other cells still showed considerable quanti- ties. On cooling the cell down to 15° C., the E.M.F. rose with a coefficient 0:00040, and then remained steady for some time, no crystals reappearing. After a time a sudden rise in the E.M.F. was observed, and the crystals were seen to have reappeared on the surface of the paste. On transferring the cells back from the melting ice to the constant temperature bath at 15° C., the E.M.F. of the B.O.T. cells was almost invariably found to be from 2 to 4 millivolts higher than before cooling. The difference was greatest in those cells which contained the greatest quantity of solution, and persisted for several days if 124 Prof. H. L. Callendar and Mr. H. T. Barnes. the cell were not disturbed. In one case, after cooling to 0° C., a cell remained nearly 5 millivolts too high when kept for one hour at 15° C. After twenty hours at 15° C., the difference still remained 3 millivolts. In the course of the next two hours it was shaken twice, and returned to within one-tenth of a millivolt of its previous value. On further shaking the cell it was noticed that, if the mer- curous sulphate were disturbed so as to come in contact with the zinc, the E.M.F. temporarily fell some 2 or 3 millivolts, but recovered very quickly on the mercurous powder subsiding. This observation illustrates the necessity, now well understood, of keeping the zinc from direct contact with the mercurous paste. § 8. Illustration of Diffusion-lag. To illustrate the extreme slowness of the diffusion process, and to show that all saturated cells have really the same coefii- cient, if sufficient time be allowed for diffusion, the following experiment was tried. Two exactly similar Board of Trade cells of normal H.M.F. were taken from the constant-tempera- ture bath at 14° C., and placed in the other bath at 25° C. Two or three observations were taken each day of their subsequent changes of E.M.F. After the lapse of two days, one of the cells was occa- sionally shaken, and rapidly gained the correct value of the H.M.F. of a saturated cell at 25°C. The other cell was left undisturbed, the temperature being maintained constant to one-fiftieth of a degree Centigrade. The H.M.F. of the latter cell fell slowly and almost uni- formly as the diffusion proceeded, but it was not until after the lapse of nearly a fortnight that it reached the correct value. The annexed curves (Fig. 1) illustrate the rate of diffusion in these cells. The abscisse represent time in days; the ordinates, fall of E.M.F. in millivolts from 15° C. For the sake of comparison, a cell of a different type, which we designate “B.O.T. crystal,” was submitted to the same treatment at the same time. The straight line BC relates.to this “crystal” cell. Tt will be seen that it shows no appreciable diffusion-lag. In fact its E.M.F. had arrived in twenty minutes within a tenth of a milli- volt of its final value. Curve No. (1) relates to the B.O.T. cell which was shaken. The points at which the shaking took place are marked by the sudden falls of H.M.F. Curve No. (2) relates to the cell which was left undisturbed. The rate of diffusion would probably have been much slower and more uniform, but for the slight vibration due to the running of the stirrer in the water-bath. The changes in the rate of diffusion shown at the points 2°5 days and 85 days were probably due to excessive stirring about those dates. Electromotive Force of different Forms of the Clark Cell. 125 Scale of Days. Gi GLO IP 45 kk 5 Fite A a Fall of EMF in Millivolts from 18°C. 45 Fie. 1.—Curves showing Diffusion-Lag of two B.O.T. Clark Cells. We have also tested several of the Muirhead portable cells in a special water-jacketed air bath, the temperature of which was regu- lated by a thermostat. We find that for comparatively rapid changes of temperature, such as 10° C. in two hours, they have a temperature- coefficient of 0°00050, on the average, between 10° and 30° C. They have also a slow diffusion-lag similar to other B.O.T. cells. It must be remembered, however, that these cells, owing to their form, are not intended for the most accurate laboratory tests, and that they are quite sufficiently constant for the purposes for which they are generally used. 3 It is sufficiently evident from the examples above given that a Clark cell of the B.O.T. form containing clear solution cannot be said to have a definite temperature coefficient. The change of E.M.F. is seen to depend on the previous history of the cell, on the rate of change of temperature, and on the quantity of solution and relative size and position of the zinc rod. ‘The temperature-coefficient at 15° C. may have any value, from 0:00040 to the value 0:00078, which is gencrally taken. If the latter value of the coefficient is regularly used, errors which are relatively considerable in accurate work may readily be made, even if the rate of change of tempera- ture is only 4° or 5° a day. There is no doubt that, provided suffi- 126 Prof. H. L. Callendar and Mr. H. T. Barnes. cient time is allowed, and sufficient crystals are present, the higher value is the more correct one to use; but we are inclined to think that many observers are not aware how extremely slow the process of diffusion really is, and how considerable a time is required for the attainment of the hmiting value. § 9. Board of Trade “* Crystal” Cells. The defects of the ordinary B.O.T. form of Clark cell in this respect of diffusion-lag have long been recognised, and various methods have been proposed to remedy the disadvantages resulting from it. Lord Rayleigh himself preferred the H-form of cell, in which the zine is buried under a layer of crystals. This form of cell has been adopted by the Berlin Reichsanstalt, and has been shown by Kahle to be practically free from the defects of the other pattern. We have made several cells of the H-form, but we are strongly of opinion that, besides being more difficult to make, they are not so convenient to work with as the B.O.T. test-tube pattern. On considering the matter in the light of the preceding obser- vations with regard to the great differences in the temperature- coefficient produced by different degrees of immersion of the zinc rod in the crystals, it occurred to us that the diffusion-lag of the B.O.T. pattern might be entirely removed by a very simple modi- fication of procedure ; so simple, in fact, that it would seem scarcely to deserve mention, were it not that of the many hundreds of B.O.T. cells which we have seen and examined, not one has been constructed in the manner to be described. The modification we have adopted in these cells, which we term Board of Trade “crystal” cells, consists simply in filling the cell above the mercurous paste with moist crystals, instead of with satu- rated solution. Under these circumstances no part of the solution can remain either supersaturated or unsaturated for any appreciable time. We have subjected these and other cells to the severest tests, and the most sudden variations of temperature, such as 0° C. to 25° or. 30° C., and we find that the B.O.T. cells, when filled in this manner with crystals, have no appreciable diffusion-lag, and are not sur- passed in quickness by any other form. § 10. Preparation of Crystal Cells. The procedure which we adopt in making these crystal cells differs only in one or two small details from that preseribed in the Board of Trade memorandum on the Clark cell. A stock solution with mercurous sulphate paste is prepared Electromotive Force of different Forms of the Clark Cell. 12% exactly as therein described. When cool, the supernatant liquid may be decanted off to be used for the preparation of crystals. A supply of suitable crystals is easily obtained at any time by taking this solution, or any other solution of zinc sulphate which has been neutralised and treated with mercurous sulphate at 30° C., and cool- ing it down to 0°C. The liquid is then decanted off, and the crystals drained on a piece of filter-paper. This method leaves them suffi- ciently moist for the purpose. The crystals are filled into the cell? through a glass tube or funnel to a depth of about 2 cm., and a hollow is made in the surface with a glass rod to facilitate the introduction of the zinc. When the cell has settled it should be free from air- bubbles, and should show the merest film of liquid above the surface of the crystals. Inasmuch as zine sulphate does not tend to form any hydrate higher than the hepta-hydrate between the limits 0° and 30° C., the crystals will remain equally moist, or very nearly so, between these limits. After soldering the platinum wire on to the zinc rod, we prefer to seal the end of the zinc rod with marine glue into a glass tube which nearly fits it. The object of this is to make the best possible seal to protect the solder joint, which may otherwise be injured by the creeping of the solution. The upper part of the glass tube is then sealed on to the platinum wire to forma mercury cup. The glass tube also forms a convenient handle to use when inserting the cork and the zinc rod into the cell. We prefer to amalyamate the zine rod, as this proceeding appears to protect the zinc from local action, and to give more uniform results. We may here remark that in the B.O.T. cells containing solution the zinc rod, even if amalga- mated, rapidly becomes corroded near the top by local action. In the cells filled with crystals, on the other hand, the zinc remains perfectly bright and clean. § 11. Decomposition of Mercurous Sulphate. After neutralising the zinc sulphate solution with zinc oxide, we prefer to filter in a jacketed funnel at 40° C., for the sake of obtain- ing a stronger solution. Mercurous sulphate is then added to remove any traces of zinc oxide or other impurities which have any action upon it. We have observed that if, after the addition of the mer- curous sulphate, the solution be heated to between 35° and 40° C., a ‘slight change may be noticed in the appearance of the mercurous sulphate. The filtrate, when cooled to 15° C., may remain clear, but, if further cooled to 0° C., a yellow turbidity makes its appearance, showing that the mercurous sulphate has, in all probability, been partly decomposed by exposure to the higher temperature. If, on the other hand, the zine solution has not been heated above 30° C. 128 Prot el. L. Callendar and Mr. H. T, Barnes. with the mercurous sulphate, the filtrate will remain clear when cooled to 0° C. Itappears probable that if a Clark cell is heated above 30° C., or is made from paste which has been so heated, its E.M.F. may be affected by a similar cause. The temperature of 30° C. appears, however, to be a perfectly safe limit. It is not unlikely that the decomposition is determined by the presence of zinc oxide, as we have rarely observed changes of more than one or two tenths of a millivolt even after heating a cell to 50° C. § 12. Tests of B.O.T. Crystal Cells. Several of these crystal cells were made by H. T. B., and later oe the advanced electrical students in the ordinary course of their work. The cells so set up at different times by different students were rarely found to differ under any conditions by so much as the tenth of a millivolt from the mean at any given temperature. If, as occasionally happened, a new cell, within an hour or so of sealing up, was found to have an H.M.F. as much as two tenths of a millivolt too high, it was short-circuited for half an hour or so with a piece of copper wire. This procedure invariably had the effect of reducing the H.M.F. to its normal value. It might naturally be supposed that with so small a quantity of solution, these cells would be seriously affected by short-circuiting. We have found, on the contrary, that they are much less affected than the ordinary B.O.T. form or than unsaturated cells. The crystal cells on short circuit were found to give a current of about 5 or 6 milliampéres, falling gradually to 2 or 1 in the course of an hour. On removing the short circuit the cells instantly recovered to within a millivolt of their normal value, so quickly, in fact, that it was found impossible by the balance method to obtain any inter- mediate readings showing the rate of recovery. In less than five minutes the value had generally recovered to within a tenth of a millivolt of the normal. The ordinary B.O.'T. cells of the same size, containing clear solu- tion, were found to give a similar current on short circuit, but the recovery was never so rapid or perfect. § 13. Temperature Coefficient of Crystal Cells. Having satisfied ourselves by various preliminary trials that the crystal cells were practically free from diffusion-lag, and finding that the temperature coefficient between 15° and 0° C. appeared to be somewhat higher than that given by Kahle and other authorities, we determined to make a systematic series of observations under definite and uniform conditions. Electromotive Force of different Forms of the Clark Cell. 129 For this test four crystal cells of different dates were selected, differing as widely as possible (one-tenth of a millivolt either way) from the mean at 15° C. Six other cells were set apart as standards of comparison, and were kept at a constant temperature of 15° C. night and day throughout the series of observations. The cells were all of the same form and dimensions as the B.O.T. cells above described, and were all sealed with marine glue. At starting the mean of the test cells at 15° C. agreed to one hundredth of a millivolt with the mean of the standard cells. The four test cells were then immersed in melting snow, and comparisons of their E.M.F. with the standards were made at intervals of six, eighteen, and twenty-four hours after immersion. For the next twenty- four hours they were kept at a temperature of 5° C., next day at 10° C., then for three days: at 15° C., then for a day each at 20°, 25°, and 30° C. In this manner the comparisons at each point were made as nearly as possible under similar conditions. The small differences which the cells possessed at starting were maintained, within two or three hundredths of a millivolt, throughout the series of observations. They were possibly due to inequalities of age, or to slight differences in the preparation of the solutions and crystals, or to the fact that the platinum wires in the mercury cups were notamalgamated. No systematic difference in their temperature- coefficients could be detected. At each point the mean of theobservations taken at the end of the first six hours showed a slight lag, as compared with the obser- vations taken at eighteen and twenty-four hours, amounting on the average to nearly two hundredths of a millivolt. It is necessary to remark, however, that a difference of one hundredth of a degree of temperature means rather more than one hundredth of a millivolt; and that this apparent lag, corresponding to less than 0°02° U., may have been due to some constant error of observation of temperature under different conditions in the morning and evening. In any case it is evident that the diffusion-lag, if any, is so small that it would be quite useless to consider it in any case, unless the greatest pre- cautions were taken to secure a constant and uniform temperature, . and to measure the temperature to at least 0°01 C. After the three days at 0°, 5°, and 10° C., the mean of the four cells returned in less than twenty-four hours to within one hundredth of a miilivolt of the mean of the standard cells. After three days at 15° C. the mean values were identical. At the conclusion of the observations at 30° C., the cells were replaced in the bath at 15° C. In fifteen hours the mean value had returned to within two hundredths of a millivolt of the standards. Some two months after the above series of experiments, a second set of observations at 30° C. was taken with three of the same cells, by 130 Prof. H. L. Callendar and Mr. H. T. Barnes. way of verification. The cells at 15° C. were still found to preserve their relative differences to within one or two hundredths of a milli- volt, and the mean still agreed with that of the standards. The difference at 30° C. was observed directly in terms of the resistance boxes, as well as in terms of the bridge-wire, at each observation. The fall of H.M.F. found was less by four hundredths of a millivolt in 19°4 millivolts, than on the preceding occasiom. After twenty- seven hours at 30° C. the cells, when replaced in the 15° C. bath, returned in twenty hours to within two hundredths of a millivolt of their previous values. § 14. Quickness of Recovery. Jt must not be hastily assumed from the foregoing observations that the cells in each case took nearly a day to recover their original values at 15° C. The observations were taken in this particular series after the lapse of several hours in order to make sure that the cells had not undergone any permanent change as the result of prolonged exposure to 0° and 30° C. We have made several special tests with regard to this point. We find that B.O.T. crystal cells of this size, set up in test-tubes nearly 2 cm. in diameter, when suddenly transferred from melting snow, or from the other bath at 30°, or even 40° C., back to the constant tem- perature bath at 15° C., return to within a tenth of a millivolt of their previous values in less than ten minutes. For cells of a smaller size, of which we have made several, the recovery is still more rapid. It appears to be chiefly a question of the time required for the change of temperature. After exposure to a temperature above 15° C., the cells frequently return, at 15° C., in less than half an hour to within one or two hundredths of a millivolt of their previous values. After exposure to a temperature below 15° C., the recovery is a little slower. This might naturally be expected, as crystallisation is generally more rapid than solution. We have also taken a series of observations at 40° C., though this temperature lies outside the limits of practical utility. On suddenly raising the temperature to 40°6° C. the value observed after ten minutes was one millivolt higher than the final value. The next observation was taken after three hours, by which time it was found, on subsequent reduction, that the H.M.F. had become constant. The E.M.F. at 40°60° C. was found to be 35°81 millivolts lower than at 15° C. The recovery on returning to the 15° bath after a day at 40°6 C. was quite unexpectedly rapid. In ten minutes the vaiue was found to be within a tenth, in two and a half hours within a hundredth of a millivolt. Later observations showed no further change. Elleetrometive Force of different Forms of the Clark Cell. 131 § 15. Results of Observations. Table I.—Temperature-variation of B.O.T. Crystal Cells. Difference in millivolts Difference from Tempera- from value at 15°C. lineality. Difference ture C. 0 SES i as rae se : formula Te ae Observed Formula | Observed | Calculated (p). St (O). (L). (O)—(L). | formula (7). 0°00° + 16°62 +18 :00 —1°38 —i°40 + 0°02 5°17 +11°15 +11°80 —0°65 —0°60 —0°05 9°89 + 5°92 + 6°13 —O°21 —0O:16 =O505 20°43 — 6°70 — 6°52 —0°18 —-0°18 —0°00 24°75 —12 +25 —11°-70 —0°55 —0°59 +0°04 29 +90 —19°42 —17°88 —1°54 —1°38 —0°'16 29 86 —19°32 = 17 63 —1°49 —1-:37 —0:12 | 40 *60 — 35°81 — 30°72 —5 ‘09 | —4°07 — —1-02 Fia. 2. - Scale of 7 CMpErature Centigrade. g° a fo° lg Zo" 25° 30° t 8 G Scale or Nilfivoles. S$ 1 z G The results of these experiments on the temperature-variation of the H.M.F. of B.O.T. crystal cells are given in the above table, and are plotted in the accompanying curve (fig. 2). The curve is drawn to show not the whole temperature-variation of the H.M.F., but the defect of the change from lineality. The observed H.M.F. in volts at a temperature ¢° C. is less at all points than that calculated by the simple linear formula, Lp We 0 00200 (S15). as oe as rete oe ICE: The difference from this linear formula is approximately repre- sented by the addition of a parabolic term, —0-0000062 (E—15)?......-- ie Re a2 MOM.) EXCL. L 132 Prot. H. L. Callendar and Mr. H. T. Barieas The full curve in fig. 2 represents this difference-term on a scale of 24 cm. to 1 millivolt. The crosses represent the results of actual observation at the different points. The mean difference of the observations from the curve, if we except those at 30° C., is only three-hundredths of a millivolt. The observations at 30° C. differ by more than one-tenth of a millivolt,and those at 40° C. by a whole millivolt from the simple parabolic curve. These differences cannot, we think, be explained as being due to errors of observation. This is proved by the accuracy with which the cells returned to their original values at 15° C., and also by the agreement of the twenty or thirty different readings at each point. Moreover, the observations have been repeatedly verified by cthers, not shown on the curve, with cells of different types, to within one or two-hundredths of a millivolt. We conclude that no simple parabolic formula can be made to fit the observations throughout the range 0° to 40° C. Over the range 0° to 28° C., however, the differences, even if real, are not of great importance, and we may take the formula, E, = E,,—0-001200 (#—15)—0-0000062 (¢—15)*,...... (L) + (P) as represcnting the temperature-variation of the H.M.F. of these cells within about one-twentieth of a millivolt between these limits. The symmetry of the curve shows that we may tuke the very convenient round number 1°200 millivolt, for the change of EH.M.F. per 1° C. at 15° C. Taking formula (P), we find for the temperature-coefficient at t° C., d/dt (H,/Ei;) = —0-000837—0-0000087 (¢—15), and for the mean temperature-coefficient between t° and 15° C., (H;/E,;;—1)/(t—15) = —0-000837 —0-0000043 (¢ — 15). Itis generally, however, more convenient to use the formula (L+P). Between the limits of 12° and 18° C. we may use the simple linear formula (lL), without risk of making an error greater than one- twentieth of a millivolt. If, however, we were to use the temperature- coefficient 0°09076 (which is very commonly taken) over the same range, the error might amount to nearly three-quarters of a millivolt. § 16. Results of Previous Observers. The formula given above for the temperature-coefficient differs from that of previous observers chiefly in the direction of making the change of E.M.F. more nearly uniform. Lord Rayleigh* tested two cells under similar conditions of slow temperature change. For one cell he found the temperature-coefii- cient, at ¢ given by the formula 0:00083+0-000018(t—15), which * ©Phil. Trans.,’ vol. 176 (1885), p. 794. Electromotive Force of different Forms of the Clark Cell. 133 agrees with our value, except that the rate of change of the coeffi- ecient is twice as great. For the other cell he found a coefticient 12 per cent. smaller with a similar rate of change. His cells, in which the zinc was pushed down into the paste, would be certainly less liable to diffusion-lag than the ordinary B.O.T. pattern, and might possibly possess a different coefficient ; but we think that the ’ difference is most probably to be explained by diffusion-lag, and in that case the higher value would be the more correct. Glazebrook and Skinner find the value 0:00076 for the mean coefficient between 0° and 15° C., under conditions in which diffusion- lag would be approximately eliminated. Our value between these limits is 0°000773. | Our value for the coefficient at 15° C., namely, 0°000837, is also in fairly close agreement with the value 0:00082 for an H-cell at 15° C., found by Fleming. The lower values obtained by many observers for ordinary B.O.T. cells, are doubtless vitiated by the effects of diffusiou-lag, and are, in this respect, in agreement with our own results for such cells under similar conditions as given in a previous section. f The observations of Kahle* are probably the most systematic. He finds for the E.M.F. in volts at ¢° C. the formula E, = E,,—0:00116 (15) —0-00001 (¢—15)? ,.... (K). The difference between this formula of Kahle and the linear formula (L) is shown by the dotted curve in fig. 2. It will be seen that the agreement with our observations is very close between 25° and 30° C., but that the value of the coefficient at 15° C. is some- what smaller. Below 10° C. the divergence is very marked. The formula of Kahle gives a change of only 15°15 millivolts between 0° and 15° C., corresponding to a mean coefficient of 0:000704, values which are evidently much smaller than those given above. We do not think, however, that it is necessary to assume that there is any real difference of behaviour between our cells and those tested by Kahle. The discrepancy is more probably to be explained by the fact that the observations on which the formula of Kahle is founded were taken between the limits 12° and 28° C., under condi- tions less favourable to the cells. Between these limits, so far as we are able to judge, we are in agreement with Kahle, within the limits of accuracy of his observations. Kahle does not give any detailed observations or any statement of the probable error of his results, but it is possible to form a general idea of the limits of accuracy from the account which he gives of his method. The cells were kept immersed. im paraffin baths, regulated by means of “ Rohrbeck” thermostats. The temperature seldom varied more than one degree from day to * “Wied, Ann.,’ vol. 51 (1894), p. 197. L 2 134 Prot. H. L. Callendar and Mr. H. T. Barnes. day in either bath, or by more than a tenth of a degree in the course- of an observation. From other observations given by Kahle* it would appear as though his cells were subject to a temperature or diffusion-lag of the order of half a millivolt, when the temperature: was changing at the rate of 1° C. per hour. From these and other considerations it is evident that Kable did not aim at an order of accuracy higher than one or two-tenths of a millivolt, and that his: formula could not be expected to give correct results beyond the limits of observation. Thus, although his formula is practicaily correct between the limits of his observations, namely 12° and 28° C., it is quite possible that it may be as much as 1U per cent. in error at 0° C. On the other hand, we regard it as quite impossible that our observations at this point should be in error by even a tenth part of the amount, namely 14 millivolts, by which they differ from the formula of Kahle. Again, although the observations of Glazebrook and Skinner between 15° and 0° C. may have been affected to a slight extent by diffusion-lag, it is plain that the effect of diffusion-lag, if any, must have been to reduce the extent of the change, and could not explain the fact that the change which they observed was so much larger’ than that given by Kahle’s formula, and so nearly in agreement with our own. In this connection it is necessary to refer to an opinion which we have often encountered in conversation and otherwise, and which is possibly still current, namely. that the observations of Glazebrook and Skinner are in precise agreement with those of Kahle on this: point. For instance, Schustert quotes correctly Kahle’s formula for the mean temperature-coefficient between ¢° and 15° C., namely, a = 0:000814+ 0:000007 (¢—15), .and states that Glazebrook and Skinner’s coefficient (a = 0:00076) refers to a mean temperature of 75° C., and is identical with the above at that temperature. This is obviously true if we put ¢ = 7:5° in the formula, but not if we put ¢ = 0°. The mistake appears to. have arisen from the above formula having been inadvertently described by Kahle as being the temperature-coeficient at t, instead of the mean coefficient between ¢ and 15° C. But although the words, “ Hiir eine beliebige Temperatur t,’ used by Kahle, may, perhaps, be ambiguous, the complete formula (K), from which the other is. derived, leaves no possible doubt as to the true meaning.{ * Loc. cit., p. 199. + ‘ Phil. Trans.,’ A, vol. 186 (1895), p. 458. t Note added Sept. 20, 1897. Ina more recent number of ‘ Wied. Ann.,’ Oct.,. 1896, Kahle states incidentally that he has found by direct comparison the dif- ference 16°6 millivolts between 0° and 15°, instead of 15°1 as given by his previous. formula. No details of observations are given. Electromotive Force of different Forms of the Clark Cell. 135 § 17. Hermetically Sealed Cells. We have always felt somewhat dissatisfied with the usual practice -of sealing up standard cells with marine glue or paraffin wax. Pro- vided that the marine glue sealing is carefully and conscientiously performed, it may, doubtless, remain good for a considerable time, if the cell is not exposed to extreme variations of temperature. We have not, however, ourselves succeeded in making the marine glue seal stand many repetitions of the 0° to 30° C. treatment. Cells which have been thus treated for a month or two have invariably shown some signs of creeping. Hxcept in extreme cases, this creep- ing does not appear to produce much effect on the H.M.F. of saturated cells, but in the case of unsaturated cells the effect is serious. The set of twelve Carhart-Clark cells in our collection, though evidently prepared and sealed with the greatest skill and are, have all suffered from the creeping out of the solution in the lapse of two years. One of the cells is now 5 millivolts below its normal value. It is only fair to add that, owing to the extremes of the Montreal climate, they have been subjected to an annual tem- perature range of 5° to 27° C., and that all the Muirhead cells in our collection are similarly affected. ‘We have succeeded in making several forms of hermetically sealed cells, and we are of opinion that such cells are much to be preferred as standards to those sealed in any other way. The following are the principal varieties on which we have made experiments. (1) Cells of the H-Type with Zinc Amalgam.—We prefer to make this cell in the form (x) of an inverted Y. Fine platinum wires are first sealed into the lower extremities of the inverted Y, the limbs are then filled with zinc amaleam and crystals of zinc sulphate, and with mercury and mercurous paste as usual. When suflicient materials have been introduced, the middle leg is sealed off. This inverted Y-form is much easier to make than the H-form, as it involves only one T-join. In making these cells, we prefer to use lead glass tubing about 5 to 8 mm. in bore. Hermetically sealed ells of this form were made many years ago by Wright* for the purpose of testing the effect of dissolved air on the E.M.F, W-Form.—When intended for immersion in a water bath, the lower limbs of the inverted Y are continued upwards beyond the seal to a height of 4 or 5 inches forming a W. The upturned limbs are partly filled with mercury, and are used for making connections. As the result of our experience with cells of this description, we are not inclined to recommend the use of cells containing zinc amalgam at temperatures above 25° C. or below 10° C. As Lord siadsinls has observed, these cells show a very remarkable tendency * © Phil. Mag.,’ vol. 16 (1883), p. 28. 136 Prof. H. L. Callendar and Mr. H. T. Barnes. to crack in the platinum seal of the leg containing the zinc amalgam, especially if exposed to low temperatures. We do not think that they could be trusted to stand the Montreal climate. Very few of our cells constructed on this pattern have survived a month or two of the 0° to 30° treatment, if they contained more than a mere button of amalgam. (2) H-Horm Cells with Zinc Rod—We have generally found cells made with zinc rod to be more reliable. The zinc rod is cast in a small glass tube of suitable size. The platinum wire, previously enclosed in a capillary glass tube, is thrust into the fused zinc. When cool, the glass mould is broken off, and the zine rod cleaned and amalgamated, and introduced into one leg of the H. The other leg is partly filled with mercury to which connection is made by a platinum wire in a glass tube after the Board of Trade method. The other materials are filled in as usual. The two legs of the H are then fused up at the top, the upper portions serving as mercury cups. We prefer to use tubes of very small dimensions, and to make the horizontal connecting limb as short as possible. This cell is not very easy to make, owing to the double fusion on to the platinum wires after the materials have been filled in, when the cell cannot be inverted. If, however, the tubes are made sufficiently small, it is a very convenient and sensitive form of cell. (3) Board of Trade Form with Crystals—The single tube form with the zinc rod cast in a similar way on to a glass capillary, is equally efficient if filled with crystals, and is much easier to seal. There is only one tube to seal, and it is possible to get at it evenly from all sides. There is generally no difficulty in keeping the two wires separate, provided that the capillaries through which they pass have been drawn sufficiently thick and strong. Both the capillaries may be expanded into mercury cups at the top, or the outer tube itself may conveniently be used to form the mercury cup for the zinc terminal. Portable Yorm.—This cell is still more easily made in a portable form, in which the mercury is replaced by amalgamated platinum. The cell may then be made upside down, the difficult seal being made first, before the materials are filled in. The process of making the cell is briefly as follows. A platinum wire is sealed into a thick glass capillary with a smal] mercury cup at one end. The wire is left projecting some 2 or 3 cm. beyond the glass at each end. One end is then hammered flat to serve as the positive element, and both ends are amalgamated by heating and plunging in mercury. A platinum wire capillary without a mercury cup is cast into a small zine rod, and the free end is amalgamated. A glass tube of suitable size and length to form:the cell is melted down in the middle till it Electromotive Force of different Forms of the Clark Cell. 137 is of sufficient size to just admit the passage of the two capillaries from opposite sides. The capillaries are then held in their proper relative positions while the seal is completed, and the free ends of the wires are coiled down in their respective mercury cups. The lower end of the tube is then drawn out slightly to facilitate the final sealing off. The materials are filled in after the usual method, but in the reverse of the usual order. Moist erystals of zinc sulphate are packed. round the zinc till it is covered to a depth of about a centi- metre. After inserting about half the mercurous paste, the flattened and amalgamated platinum wire is coiled down into a suitable posi- tion, and more paste is added. The end of the cell close to the point where it is to be finally sealed, is preferably filled with moist crys- tals instead of paste. The object of this is to avoid leaving any of the paste close to the seal, where it might suffer decomposition from the heat in sealing off. Any excess of solution is dried off with filter paper, and the narrow neck is then sealed off with a fine flame. On the whole, we prefer this portable form of cell to any of the other forms we have tried. The shape of the cell and its small size make it very convenient. The seal is comparatively easy to make, and the narrow glass neck separating the cell from its connexions is also an advantage, as it diminishes the risk of any error of tempera- ture arising from conduction along the tube. We have not found that there is any advantage in using mercury as compared with amalgamated platinum. The cell has a higher internal resistance, and gives a smaller current on short circuit, but the recovery appears to be equally rapid and complete. We cannot find any systematic difference in the electromotive force at any temperature. If anything, the amalgamated platinum has the advan- tage over the mercury, as the purity of the mercury is then compara- tively unimportant, and redistillation is unnecessary. There does not appear to be any advantage gained, in our experience, by using a strip of platinum foil in place of the fine flattened wire. § 18. Tests of Hermetically Sealed Cells. We have similarly tested several saturated cells of the H-form, and various other patterns above described as hermetically sealed cells. We find that they all show the same temperature change of H.M.F. as the B.O.T. crystal cells. The agreement in nearly every case is within one or two-hundredths of a millivolt even at O° and 30° C., the limits of the range. The largest divergence was found in the case of a portable cell of type (3), with an amalgamated platinum wire in place of mercury. In this cell, both at O° and 30° C., the difference from 15° C. 138 Prof, H. L. Callendar and Mr. H. T. Barnes. exceeded by one-tenth of a millivolt that of all the other erystal-cells we have tested. Even in this case the cell was at least consistent with itself. It was first tested at 15°, 30°, 15°, 0°, and 15° C., allow- ing a day at each temperature. It was a new cell, the first of this type which we tested, and its value at the time was nearly three- tenths of a millivolt higher than the standard. The value found at 0° C. was 16°72 millivolts above, and at 30° C. 19°72 millivolts below its value at 15° C., instead of +16°62, and —19°58 millivolts respec- tively. The cell being of a very small and sensitive form, the test was repeated a few days later in the reverse order, allowing only half an hour at each temperature. The values found in the second test were +16°71 and —19°70 millivolts respectively. We thought at first that the discrepancy might be due to some inherent peculi- arity of this type of cell. We have since tested other cells of the same type and size, with results which agree to 0°02 millivolt with the B.O.T. crystal cells. The agreement is not confined to points 0° and 30° C. For instance, the difference found at 24° C. was 0°05 of a millivolt less than that calculated by the formula (P). The B.O.T. crystal cells at this point show a difference of 0°04 millivolt from the curve (P) in the same direction. We consider that the divergence of one-tenth of a millivolt in the case of this particular cell must be regarded as exceptional. § 19. Importance of Constant Conditions of Temperature. We have quoted the above test partly as an illustration of the kind of agreement between cells of different types which it is possible to attain with suitable cells under definitely known conditions of tem- perature. In attaining this order of accuracy the chief difficulty lies in the certain determination of the temperature of the cells. To attain a certainty of the order of 0:01° C., the following conditions are necessary :— 1. The cells should be of an elongated form, and should be deeply immersed in a bath of liquid, which is constantiy and vigorously stirred. 2. The temperature must be kept constant to 0-01° C., and the thermometer used must be read and corrected to the same order. If the cells are. not deeply immersed their temperature will be affected by external conditions. If the liquid is not constantly stirred it will tend to be hotter at the upper surface, especially if the liquid be very expansible, like paraffin, and the bath be hotter or colder than its surroundings. If the temperature is changing the cells will not be of an uniform temperature throughout, and will lag behind the thermometer, unless they happen to be of a smaller size. Electromotive Force of different Forms of the Clark Cell. 18% The electrical conditions are not less important, but are more easily realised and maintained. Our galvanometer is sensitive to much less than a millimetre of the bridge-wire (one hundredth of a millivolt). Great attention is paid to the perfection of the insulation, and to the avoidance of thermo-electric effects, which may readily amount to more than ten microvolts. We may here remark that in testing any new batch of cells it is quite impossible to tell, till the results are worked out, whether they are in agreement with others. The many coincidences found cannot therefore be the result of bias on the part of the observer. We think we may fairly claim for the Clark cell an order of con- sistency approaching one-hundredth of a millivolt in the temperature changes of its H.M.F. § 20. Clark Cells in which the Solution is of Constant Strength. It is well known that Clark cells, not containing crystals, in which the solution does not change its strength with change of temperature, have the advantage of possessing a temperature-coefficient which is less than half that of the saturated cells. They are also practically free from the effects of diffusion-lag, as the density of the solution is always nearly uniform. The best known cell of this type is the Carhart-Clark cell, in which the zine sulphate solution is chosen as being saturated at 0° C. An error of 2° C. in the temperature at which the solution is saturated, will make an error of only one millivolt, approximately, in the H.M.F. of the cell. We have prepared several cells of this type at different dates and in different forms, with separately prepared solutions. In cells so prepared, of similar patterns, we have not as a rule found differences greater than two or three-tenths of a millivolt. These differences were probably due as much to other cause as to difference of strength of solution. We have generally sealed the cells hermetically to avoid creeping of the solution, which has a tendency to lower the E.M.F. in the case of unsaturated cells. It is evident that these cells must, on the whole, be less accurately reproducible than the saturated cells. We have also found that they are more liable to undergo slight changes of E.M.F. as a result of Short-cireuiting, or of exposure to high or low temperatures. They appear, in fact, to be less stable than the cells containing crystals. We have also prepared experimental cells with solutions weaker than the cell saturated at 0° C. We have observed in these cells a similar instability, becoming more marked as the solution is weakened. Weare inclined to attribute this instability to a difference in solubility or diffusivity of the mercurous sulphate in the weaker solutions. 140 Prof. H. L. Callendar and Mr. H. T. Barnes. We also found it possible to obtain observations of the change of E.M.F. of cells saturated at 15° and 30° C., under conditions in whicls the solution was considerably supersaturated. The cell saturated at id5° C. was kept for several hours at 0° C. without showing any trace of crystallisation. This cell agreed at 15° C. with the saturated cells, and gave very consistent readings throughout the range 0° to 30° C.. The change of E.M.F. per 1° C. was found to be 0°567 of a millivolt between 0° and 15° C., and 0°560 of a millivolt between 15° and 30°. As the whole change of E.M.F. between 0° and 50° C. was only 16°90: millivolts, and as the E.M.F. of the cell rose by one-tenth of a milli- volt after keeping for six hours at 0° C., the observations may be taken as showing that the temperature-coefficient of this cell, whether in the supersaturated or unsaturated condition, is practically constant over the range 0° to 30°C. This is in marked contrast with the case of the cells containing crystals. The tests of the cells saturated at 0° C. were very fairly consistent, but not quite so good as those of the cells saturated at 15° C. They showed a mean temperature change of E.M.F. per 1° C. of 0°543 of a millivolt. ‘There was no decided evidence of any variation of the temperature-coefficient over the range 0° to 30°C. Carhart gives the formula :— | E, = E,; [1 — 1-000387 (15) +0:0000005 (t—15)?], which would make the temperature-coefficient diminish slightly as the temperature rises. (Change of H.M.F. 0°56 my. per 1°C.). The tests on the weaker cells were much less consistent, owing to the instability of H.M.F. above referred to. The results of the tests. pointed to a mean change of 0°55 of a millivolt per 1° C. The change observed between 0° and 15° C. was sometimes greater and sometimes: less than that between 15° and 30° C., but there was no decided ten- dency either way. After keeping for some time at 0° or 30° C. these: weak cells sometimes showed permanent changes amounting to as much as half a millivolt. § 21. On the Density of Solutions of Zine Sulphate. A knowledge of the density of solutions of zine sulphate is required in order to trace the relation between the changes of E.M.F., which depend on change of strength and density of the solution. This point has been investigated by two of Professor Carhart’s students, the result of whose work has been published by Professor Carhart.* These observers find for a cell saturated at 15° C. an H.M.F. nearly five millivolts higher than that of a cell containing crystals, and a density which appears to be correspondingly low. For this and * “Proc. Amer. Elect. Eng.,’ 1892, p. 615. Electromotive Force of different Forms of the Clark Cell. 141 other reasons, we thought it would be desirable to repeat the deter- minations. The observations of Lannoy,* though evidently under- taken with great care, did not extend to the case of solutions as dense as those occurring in Clark cells. Other observers appear to have confined themselves chiefly to the case of very dilute solu- tions. According to the views of Valson and Bender, which are quoted by Ostwald and other authorities, the density of a salt solution, such as zine sulphate, may be additively deduced from the observed den- sities in the case of some standard solution (e.g., a solution of ammonium chloride), by means of two moduli representing the acid and the base respectively. We have calculated the densities accord- ing to the values which they give for the moduli. at 18° C., but it appears that the results are only a rough approximation, and miss what seems to be the most characteristic feature of the change of density. In determining the relation between density and strength of solution, the chief difficulties to be encountered are in the exact measurement of the strength. If the composition of the solution is determined by weighing out known quantities of the hydrated salt into a litre flask, it is very possible that errors may arise from evaporation or efflorescence, or from the presence of other hydrates. In order to avoid these possible errors, we adopted the much more laborious method of evaporating a known weight of solution to dryness at 100° C., assuming that the residue was the monohydrate. Two determinations were made in this manner for each solution tested, and in addition, two control experiments were made in which the strength of the solution was measured by estimating the sulphate by means of barium chloride. The following table contains the results of these determinations for seven different solutions. Table I1.—Density of Zinc Sulphate Solutions. fi | — | ; a! : - Density ZnSO, Difference Calculated Ealstion of solution gram per ¢.c¢. 0 -9982 | by (D) (p.) : (d). pd/100. +w—d, formula. 6°35 1 °0653 0 -0677 0 -0006 0 0000 8°46 1 -0896 0 0923 0 -0009 90-0000 13°49 1°1522 0°1557 0 °0017 0 :0006 17°69 1°2020 0°2130 0 0092 0 0070 23°75 12872 0 3062 0°0172 0°0174 Py lr 1°3418 0 +3667 0 °0231 0 0242 33°21 1 °4400 0 °4790 0:0372 0 °0367 * Ostwald, ‘ Zeit. Phys. Chem.,’ Nov., 1895. 142 Prof. H. L. Callendar and Mr. H. T. Barnes. The first column contains the percentage by weight, p, of ZnSO, in grams per 100 grams of solution, as deduced from the observa- tions on each solution. The second column contains the values of the density, d, at 20° C., obtained by weighing a special form of pipette carefully filled with the solution. The third column gives w, the weight of ZnSO, in grams per cubic centimetre of solution, and is obtained by dividing the product of the numbers in the first two columns by 100. If we add to this weight w, the number 0°9982, representing the weight of water in grams per e.c. at 20° C., and subtract the observed density, d, of the solution, we obtain as the difference given in the fourth column, the weight of water displaced per c.c. by the zinc sulphate in solution. The observations of Lannoy reduced on a similar plan are as follows :— | | meets Re Difference | (p). | (d) at 15° C. pd) 100. 0:9992 + w—d. Calculated. GS a | 2 +25 1 -0226 0°0230 —00004 0 -0000 | 5 *60 1:0596 | 00-0594 ~0-0010 0-0000 11-21 11238 0°1250 +0:0004 0 -0000 16°85 1 1949 0°2013 | +0°0056 +0°0057 i These observations, taken in conjunction with our own, would appear to indicate a simple relation between the density and the composition, of a kind which so far as we are aware has not been pre- viously observed. Upto a density of about 1:150, the solution of zinc sulphate appears to take place approximately without change of volume. The added molecules of ZnSO, do not appear to displace any of the molecules of water, so that the density at 20° C. is very nearly 0°9982 + w. Beyond this point, it appears that each added molecule of ZnSO, displaces one molecule of water, so that the density of the solution is very approximately represented by the expression d = 0:9982+w—18(w—0'150)/161 ...... epelD), The nature of this relation is perhaps more clearly shown by the curves given in fig. 3. In this figure, the values of ware taken as abscisse, and the corresponding values of the difference 0°9982 + w—d,as ordinates. The sharp break which occurs at the point w = 0'159 is very clearly shown both by the observations of Lannoy, which are represented by crosses, and by our own which are represented by the dots enclosed in circles. Those of Lannoy unfortunately do not extend far enough to afford a satisfactory veri- fication throughout the range, but we have no reason to distrust our Electromotive Force of different Forms of the Clark Cell. 143 own observations at the higher points, as they were all carefully verified. The dotted curve, which is practically a straight line, represents the formula of Bender and Valson, which smooths out the break. Tt is not theoretically improbable that a simple relation of this. type should be found to hold approximately in the case of salt solu- tions. At the same time it is hardly to be expected that any such simple expression should represent accurately the changes of density at all temperatures. The expansion of water is anomalous in the neighbourhood of the freezing point, and the coefficients of expan- sion of solutions differ considerably from that of water at low tem- peratures, and generally increase with increase of strength of solution. These variations in the coefficients of expansion may well introduce. secondary effects of a corresponding order in the changes of density. In comparing the observations of Lannoy with our own, which were taken at a slightly different temperature, it would appear not improbable that systematic differences of this kind may exist, but the point obviously requires much more careful investigation, as the differences shown are so small, and might readily be explained by errors of observation. For instance, at the two lowest points the density according to Lannoy is greater than 0°9992 + w. Since he apparently determined the composition of the solution by weighing out quantities of the heptahydrate, the discrepancy might be explained by a slight degree of efflorescence of the sample used for these deter- minations. In the table of densities given by Carhart, the com- position of the solution in each case is stated in terms of the per- centage of ZnSO, by weight in 100 parts of solution. If we assume, in the absence of any definite statement, that the symbol ZnSO, stands in this case also for the heptahydrate, we find that the den- sities which he gives are much greater than those found by Lannoy,, or by ourselves, the value of the density at 45 per cent. of the hepta-. hydrate, according to Carhart, being 1°345, instead of 1:318, as given by our observations. It is possible that the sample used in this case may have consisted largely of the hexahydrate, or the dis- crepancy may be due to other causes. If we take older determinations of the density of zine sulphate solutions, such as those of Gerlach or of Schiff (1559), we find that they show a general agreement with our observations rather than with the formula of Bender, but that the characteristic point to. which we have drawn attention is neatly smoothed out in the tables. which they give as deduced from the results of their observations. The point in question would not be noticed at all unless the obser- vations were plotted by the method of differences, as shown in Fig. 3, and even in that case it might readily be mistaken for an error of observation, unless the points were numerous, and had been inde- a4 Prof. H. L. Callendar and Mr. H. T. Barnes. pendently checked. If we had ourselves foreseen a simple relation of this character, we should have taken even greater pains in verify- ing the observations about this point. As it is, we hope shortly to be able to investigate the subject further, and in particular to endeavour to find similar relations in the case of other solutions. Fig. 3. © FA b orf Bender. (dotted) eS © Calendar & Barnes. 2 x Lasoy. eee «| — a f +2 “5 of “5 Abscissae . Values of w= pda/i00. DIFErENCE ‘9982+ W-a. Sg ° < § 22. Change of H.M.F. with Strength of Solution at Constant Temperature. A number of Clark cells of the B.O.T. pattern, but without crystals, were set up with the solutions above described, of which the density and composition had been carefully determined. Due precautions were taken in each case to avoid evaporation. The difference of E.M.F. from the standard at 15° C., and also at other temperatures, was carefully determined in the case of each of these cells. On plotting the results, we could not find any simple relation between the change of.H.M.F.. and the density or the percentage strength of the solutions. But on expressing the observations in terms of w, the weight of ZnSO, per c.c., and not per gram, of solu- tion, we found that the values of dE, the difference of E.M.F. from the standard at 15° C., fell very nearly on a straight line, represented by the formula :— dH = 42:0—88-0w (millivolts). The following table contains the observations for each solution tested :— _——— =—- <=?” aT Te ee Electromotive Force of different Forms of the Clark Cell. 145 Table III.—Change of E.M.F. with Strength of Solution at 15° C. T £ cell (w) ad dB Difference, ype of ce | gram per c.c. | millivolts. calculated. obs. — cale. 0°105 33°2 | 32°8 + 0°4 B.O.T. QeEEG |e | 31°6 | 31°8 | —0°2 unsaturated 0-199 | 24°6 24 °5 +0°1 0°263 19 *1 18 °9 +0°2 | Sat.atO°C...) o-401. | 65 | 6°70.Ni | etho-2 | Sat. at 15° ..| 0°478 0-0 | 0-0 0-0 | | The differences given in the last column are of the same order as the accidental changes of H.M.F. observed in the case of these weaker cells. It would therefore appear probable that in this type of cell the diminution of E.M.F. is simply proportional to the volume con- centration of the salt. In comparing the above: results, a curious point remains to be noticed. Taking a cell saturated at 15° C., the increase of H.M.F., on cooling down to 0° C., has been shown above to be 8°4 millivolts, if there is no change of strength of the solution. The increase of H.M.F., due to change of strength of solution from saturation at 15°C. to saturation at 0° C., has been found to be 6°5 millivolts. We might, therefore, naturally expect the total effect due to both causes com- bined to be 149 millivolts, whereas the saturated crystal cells, in which both causes are operative, show an increase of E.M.F. of 16°6 millivolts. The explanation of this apparent discrepancy is to be found probably in the lowering of E.M.F., due to the greater diffusivity of the mercurous sulphate in the weaker solutions. In the saturated erystal cells this diffusion is practically prevented by the dense layer of crystals. In order to test this hypothesis, some weaker cells were set up in the W form, in which the possibilities of diffusion are dimi- nished by the smallness of the tube and the increased distance between the electrodes. These cells showed, as was expected, higher values of the H.M.F. than those given by the formula, the difference amounting in some cases to between 2 and 3 millivolts. § 23. Onthe Solubility of Zinc Sulphate. Tt is well known that zine sulphate forms various hydrates, which may be obtained by crystallisation at different temperatures. These hydrates differ in point of solubility, and it is important for Clark cells to employ the heptahydrate, which has the lowest solubility at 146 Prof. H. L. Callendar and Mr. H. T. Barnes. temperatures between 0° and 30° C., and may therefore be considered the normal hydrate at these temperatures. The peculiarities cf the curve representing the temperature-varia- tion of the E.M.F. of Clark cells are undoubtedly due in the main to the very considerable changes of solubility of this hydrate with rise of temperature, and are closely associated with the formation of different hydrates at higher temperatures. We therefore thought that it would be of interest to investigate this point more closely, at least within the range of temperature over which accurate measure- ments of the H.M.F. may be readily obtained. The solubility of zine sulphate has been studied by Etard,* who discovered that the solubility of the sulphates decreased with rise of temperature at higher points of the scale. His results are expressed in terms of the percentage weight, p, dissolved in 100 parts of the soiution at a temperature t° C. Expressed in this manner, he finds that the curve representing the solubility is a straight line, the equation of which for zinc sulphate is given as bemg p = 27-6.4 026042 ..ccames eee) This equation is given as representing the solubility up to a tempera- ture of about 80° C., above which the solubility decreases. Roscoe and Schorlemmer,+ on the other hand, give the following table of solubility expressed in terms of the weight of ZnSO, dis- solved in 100 parts by weight of water. For the sake of comparison, we have reduced their results to the corresponding percentage, p, of solution, and have added two lines of results calculated from the formula of Etard, and from the observations of Poggiale. Table IV Marea iby of Zine Sulphate (Roscoe and Schorlemmer, Etard and Poggiale). Hlemiperaiire (isd scsi «'s nyat 0° 20° 50° 75° ZnSO, in 100 of water (R. &8.). 41°3 53 ‘0 66 °9 80 °4 Percentage of f R. & S. ve] 292 34 °6 40-1 44.°5 iain |p) Etard.. pitiless 27 °6 32°8 40 °6 47-1 oo P) | Poggiale .......| 30-0 34°7 40 °7 45-0 It is evident from the discrepancies shown in the last three lines that the matter requires further investigation. From our own experiments we find that the rate of diffusion in these extremely dense and viscous solutions is so slow that it is much more difficult * Compt. Rend.,’ vol. 106 (1888), p. 206. ¢ Vol. 2, PartiI, p. 262. Electromotive Force of different Forms of the Clark Cell. 147 to obtain correct results than might at first sight be expected. In particular, we find it extremely important to maintain the solution for a considerable time with continual stirring at each point at a perfectly constant temperature. This essential condition could be readily secured by the aid of the constant temperature baths already described. The method which we adopted at each temperature was to prepare a saturated solution with repeated stirring in a large test-tube in the constant temperature bath. This solution, mixed with crystals, was maintained for several hours at a constant temperature. Samples of solution free from crystals were drawn off in a special pipette at intervals and weighed. They were then evaporated to dryness at 100° C., and the percentage of ZnSO, in each case was calculated, assuming the residue to be the monohydrate. The different samples at each temperature always agreed very closely, showing the solution to have been saturated and free from crystals. The results of our observations are contained in the following table. With the exception of the iast four lines, which represent single observations, each line is the mean of two or three determina- tions. Observations in different lines were taken on different days and with different samples of solution. Table ie nee ond Barnes) of Zine Sulphate (Callendar and Barnes). Temperature Percentage of Calculated by Difference centigrade. solution. formula (see Fig. IV). (Z). (p). 29 5 + 0-270. | Obs.— Cale. 0° 29 *43 29 ‘50 0°07 0) 29°53 29°50 +0-08 0 29 -4Y 29°50 —0°01 15 -00° 33°66 33°55 +0°11 15 ‘88 33°85 38°98.) | +0°07 30°70 38 °46 37°80 + 0°66 39 "92 | 41°36 40°28 +1 °08 39°95 41 °37* 40 °29 + 1°08 40°73 41°43 40°50 +0°93 41-49 41°70 40 ‘70 + 1:00 46°40 42°68 42°02 +0 °66 49°97 | 43 °51* 43°00 +0°51 49°99 43°41 43 ‘00 + 0°41 50°00 43 *50* 43°00 +0°50 50°20 42 51 43 *O5 + 0°46 The observations marked with an asterisk were obtained by heating the solution to a temperature ten or twenty degrees above that of the bath and then allowing it to cool down to the bath tem- perature with constant stirring. VOW. LXTI. M 148 Prof. H. L. Callendar and Mr. H. T. Barnes. In order to exhibit the nature of these results on a convenient scale, it is necessary to plot, not the whole percentage, as is usually the case, but the small differences from the linear formula given in the fourth column. It is then apparent that there is a singular point in the curve of solubility at a temperature of 39°C. At this point, as the temperature rises, the rapidly increasing solubility of the heptahydrate begins to exceed that of the hexahydrate, which is also increasing, but less rapidly. At temperatures above 39°, a solution containing only crystals of the heptahydrate becomes super- saturated with respect to the hexahydrate, so that if any crystals of the latter are formed or introduced, there will be rapid erystallisa- tion, and the strength of the solution will diminish. At 39° C. the solubilities of the two hydrates are equal and amount to 41-1 per cent. of the solution; but whereas the rate of increase of the solu- bility of the heptahydrate is 0°33 per cent. per 1° C., that of the hexahydrate is only 0°22 per cent. per 1°. The crosses marked R. and S. in fig. 4, at the points 0° and 20°, correspond to the values given by Roscoe and Schorlemmer plotted on the same scale. The remaining observations of Roscoe and Schorlemmer, and also those of Etard, differ so much from ours that Hincaete (aace| aan] ea Po | 7 [Linc Forma Pe | pare A | re" iis a — Difference 4. they could not be included in the lmits of the page. The observa- tions of Poggiale do not show the diminution of solubility above 80° C. which Etard gives. There is a general agreement in the results quoted, especially between those of Roscoe and Schorlemmer and Poggiale, but if the differences may be taken as indication of the order of accuracy attained, it is evident that the results of these observers could not be expected to show the critical point at 39° C. By interpolation on the curve given in fig. 4, we find the following values for the solability as given by our observations. Electromotive Force of different Forms of the Clark Cell. 149 Temperature C. . : 0° 10° 20° 30° 40° 50° Per cent. of sol. re 29°50 | 32°24 | 35°13 | 38°22 | 41°33 | 48°45 ZnSO, in 100 aq......| 41°85 | 47°58 | 54°16 | 61°86 | 70°44 | 76°84 Compare Poggiale* ...| 43°02 | 48°36 | 53°13 | 58°40 | 63°52 | 68°75 Difterence.....-- +. —1'17 | —0°78 | +1°08 | +3-46 | +6:92 | +8:09 § 24. Change of H.M.F. at Higher Temperatures. (Between 30° and 50° C.) Finding that there was a change in the continuity of the curve representing the temperature variation of the H.M.F. at temperatures above 41° C., we decided to investigate this point more closely, although the temperatures in question lie beyond the range of the practical use of Clark cells. By a slight modification of the heating arrangements, the constant temperature baths were enabled to reach steady tempe- ratures up to and beyond 50°C. The following series of observations were taken with several cells of different types, in a manner similar to that which has been already described. Hach line represents the mean of the different cells at each point. The observations given in different lines were taken on different days, in the order in which they are given in the table. An interval of several months inter- vened between the first and the last half of the table. Table VI.—Change of H.M.F. between 30° and 50° C. Difference in millivolts from 15° C. Difference from Pea poiure linear formula centigrade. Observed. Formula (L). 40 -60° 35 ‘81 30°72 5°09 30°14 19°78 18°17 1°61 30°32 20°00 18°38 1°62 35°44 27 °39 24°53 2°86 42°58 36°56 33 ‘10 3°46 46°74 41°14 38 ‘09 3°05 48 °58 43 *50 40 °30 3°20 35°79 . 2008 24°95 3°13 40 -09 34°99 30°11 4°88 42°79 36°76 33°35 3°41 44°70 38 °88 35 “64 3°24 41°54 35°45 ; 31°85 3°60 * “Watt's Dictionary of Chemistry.’ Muir and Morley, vol. 4, p. 581. M 2 150 Prof. H. L. Callendar and Mr. H. T. Barnes. The differences from the linear formula given in the last column of Table VI are shown by the crosses on the curve in fig.5, This figure is plotted, as in the previous case, to exhibit not the whole change of the E.M.F., but only the defect of the change from lineality. For instance, at 39° C., the E.M.F. does not suddenly begin to increase, but continues to diminish at a slower rate. It was evident on plotting these observations that there was a break in the curve, and that the observations after 41° C. belonged to a different branch. To investigate this poimt more closely, a con- tinuous set of readings was taken on a pair of very sensitive cells. Starting at a temperature of 35° C., the temperature of the bath was Hia. 5. Scale of Temperature Centigrade. (e) fo) ° 30 40 50 [Fin Millivolts tar formitla (L). ~~ J Scale of Millvoles. very slowly and continuously raised by disconnecting the regulator, and readings were taken of the temperature and H.M.F. of the cells alternately every few minutes. The EH.M.F. of the cells was observed to fall at a steady rate, accurately following the curve already found for the observations at steady temperatures, until a temperature of 424° was reached. At this point there was a sudden increase of more than 2 millivolts in the H.M.F., due to rapid crystallisation of the hexahydrate, and in less than five minutes the cells had reached a point on the other branch of the curve. After remaining at this point for some hours, the bath was slowly and continuously cooled, observations being taken in the same manner during the cooling. As had been expected, the cells were found to follow the branch of the curve shown in fig. 5, corresponding to the solubility of the hexahydrate. The solution remained supersaturated with respect to the heptahydrate until a temperature of 31° was reached, at which point the H.M.F. had fallen nearly five millivolts below the normal. At this point there was a sudden crystallisation of the heptahydrate, and the H.M.F. rose in a few minutes to its normal value. It will be observed that the two branches of the curve Electromotive Force of different Forms of the Clark Cell, 151 eross at 38'8°, which is in practical agreement with the temperature of equal solubility of the two hydrates, as determined by the observa- tions on the change of solubility. § 25. Permanence and Reproducibility of B.O.T. Crystal Cells. It is now more than a year since the completion of the experiments described in the preceding pages, and we are able to add the results of more recent comparisons of the cells as evidence of the order of their permanence and reproducibility. Of the original crystal cells, made more than two years ago, we still have a few remaining. None of the cells of this type have shown any signs of failure, in spite of the treatment to which they have been subjected, but many of them have been taken away by the students who made them. Table VII.—Comparisons of Crystal Cells at 15° C. Differences from mean of cells in millivolts. Number of cell. Dec. 10, 1895.| Nov. 28, 1896.| Dec. 19, 1896.) Feb. 8, 1897.) Aug. 2, 1897. es Se <1 —0-03 —0°04 —0-08 —0-02 —0 ‘04 X 2 —0°05 +0°02 +0:06 —0-04 +0:02 xX 3 +0°10 +0°04 + 0°05 ei +0°12 : x5 +0°03 —0:08 —0°07 #0702 9 }h—0-127):2 X 6 +0°08 +0:07 +0°08 +0 °04 ie %& 10 —0-07 +004 +0°01 if £0 °10))) % 11 —0-:08 —0°06 —0°07 a: 020203 As there is no particular reason why these cells should be less permanent than other Clark cells of the B.O.T. type, the above wiil probably be sufficient proof of permanence. It will be observed that. the average difference from the mean in eacli case is nearly one- twentieth of a millivolt, The extreme difference is one-tenth. We have observed that this is about the order of agreement generally attainable with Clark cells set up at different times. Over comparatively short intervals of time, such as one month, it would appear from the above list, and from other tests, that the average change in the value of any one cell, as compared with the mean, may be expected not to exceed two or three hundredths of a millivolt, but for louger periods, such as a year, the mean change reaches one-twentieth. As a further illustration of the reproducibility of these cells. and of the close agreement in the temperature variation of the H.M.F., under somewhat exacting conditions of testing, we add a list of the 152 lectromotive Force of different Forms of the Clark Cell. cells made by the fourth-year class of electrical students during the session of 1896-1897. These cells were all of the portable B.O.T. form, with a flattened and amalgamated platinum wire in place of mercury. They were set up in test-tubes, filled with crystals, and sealed with marine glue, and were otherwise exactly similar to the erystal cells described in § 10. With the exception of those marked with an asterisk, the observa- tions were all taken by the students themselves in the course of an afternoon’s work. Readings were taken at the points 15°, 0°, 15°, 30°, 15°, allowing only about half an hour at each temperature. Table VIII.—Portable Crystal Cells, made by Students. Difference | Change of E.M.F.; Mean Date Date from %y, change Name of student.| when when Me Sasa made tested. sn a om ° ° fo} saa : at 15°. | 0°—15°. |15°—30°.| degree. Stovel..si «ios cc | Mar 2.) Mar 4.2) Oxy 16°57 19 -39* 1-199 Thomson ......| Mar.2..| Mar. 4.. —O°'11 16°60 19°38 1:199 Blair ........-.| Feb. 19 .| Heb. 21.) +0°05 9) 1GsGan akan 1°201 Burnham ......) Dee. 8.2) dan, 19%) 70702 is 157 19 :30* 1:196 Davidson ......| dam. 12.2 \\fans 19°71) 4-Orrk 16°62 19 -40 1-201 Edwards ......| Dec. 10.| Jan. 26 - +0°07 16 °60* | 19 °36* 1°199 Machean ¢..i.. 4) Deexs' <2 Dees. —0 04 16°62 19°36 1°199 McDonald, P.W.| Jan.14 | Jan.19 .| +0°14 16°63 19 *49 1°204. McDonald, J. E. | Feb. 19 .| Feb. 21 .| —0°15 16°58 19°30 1:196 Packard......é0)/ Jan. 12 Jan. 19. +0°14 16°56 19°43 1°200 atbelser® (og eon Dee. 10... | Tans ZG ¢ + 0°05 16°58* | 19°53 1° 204. Walters vo. os. | Doe. 3. Dees. s. —0°13 16°61 19 °44, 1°202 Mean of students’ cells ...... —0:005 | 16°608 | 19°400 1 °2000 It should be observed that the cells were, in most cases, tested rather too soon after being sealed up. In the course of a week or two, they were usually found to have settled down into closer agree- - ment with the standard. When kept for a longer time than half an nour at 30° C., they showed a slightly greater change of E.M.F. at this point. . Proceedings. Bites i538 November 18, 1897. The LORD LISTER, F.R.C.S., D.C.L., President, in the Chair. Dr. J. S. Haldane, Mr. George Murray, Professor H. Alleyne Nicholson, and Professor H. H. Turner were admitted into the Society. A List of the Presents received was laid on the table, and Seni ordered for them. The President reported that during the recess, at a Court held at Windsor on July 15, 1897, he, accompanied by Sir J. Hvans, Treasurer; Professor M. Foster, Secretary; Professor A. W. Riicker, Secretary ; Professor R. B. Clifton, Sir W. Huggins, and Mr. W. T. Thisel- ton-Dyer, Vice-Presidents ; Sir J. D. Hooker, Lord Kelvin, and Sir G. G. Stokes, Past-Presidents, had the privilege of being admitted to the presence of the Throne and the honour of presenting to Her Gracious Majesty the loyal Address which had been approved by the Fellows at their meeting of May 20, and that Her Majesty in accepting the Address made a gracious reply. The Address and Royal Reply are as follows :— “To the Queen’s Most Hucellent Majesty. “May IT PLEASE youR Magszgsty! “ We, your loyal and dutiful subjects, the President, Council, and Fellows of the Royal Society of London, humbly beg leave to offer to your Most Gracious Majesty, the beloved Patron of our Society, our respectful and sincere congratulations on the happy conclusion of the sixtieth year of your Most Gracious Majesty’s reign, a reign which has extended over a longer period than has that of any of your Majesty’s illustrious Predecessors. Not in the number of years only has your Majesty’s reign surpassed all others: it has also been marked by unexampled prosperity and progress. Conspicuous in that progress, and an acknowledged cause of that prosperity, has been the marvellous advancement during your Majesty’s reign of that Natural Knowledge to promote which your Majesty’s illustrious Predecessor, King Charies IJ, founded the Royal Society. Your humble and devoted servants, the Fellows of the Royal Society, feel that they, in common with all men of science, have especial reason to rejoice in the long continuance of your Majesty’s beneficent rule. VOL. UXIL. N 154 Proceedings. ““'We beg your Majesty graciously to receive this token of the loyal devotion which the men of science of this country feel towards their Sovereign Lady the Queen. ‘And we shall ever pray that your Most Gracious Majesty may be spared for many years to reign over your dutiful and loving people.” Her Majesty's Repiy. “| thank you for your loyal and dutiful Address. I am much gratified by the attachment which your ancient and learned Society expresses to My Throne and Person. I am fully sensible how far the labours and ingenuity of men of science, whom you worthily represent, have advanced the industrial and social prosperity of My people, and have tended alike to their good and refinement, and I confidently expect the same excellent fruit in years to come from the indefatigable and reverent Investiga- tion of Nature for the promotion of which the Royal Society was founded.” : In pursuance of the Statutes, notice of the ensuing Anniversary Meeting was given from the. Chair. Mr. Horace Brown, Sir Douglas Galton, and Dr. T. E. Thorpe were by ballot elected Auditors of the Treasurer’s accounts on the part of the Society. The Secretary read the following list of Papers received since the last meeting and published in accordance with the Standing Orders :— Professor J. A. Fiemine and Professor J. Dewar. Further Observa- tions on the Dielectric Constants of Frozen Electrolytes at and above the Temperature of Liquid Air. : Professor J. Dewar and Professor J. A. Ftpmine. On the Dielectric Constants of Certain Organic Bodies at and below the Tempera- ture of Liquid Air. Professor J. Dewar and Professor J. A. Ftemine. On the Dielectric Constants of Metallic Oxides dissolved or suspended in Ice cooled to the Temperature of Liquid Air. Professor Russert Boyce and Professor W. A. Hurpman. On a Green Leucocytosis in Oysters associated with the Presence of Copper in the Leucocytes. Cuartes S. Tomes. On the Development of Marsupial and other Tubular Enamels, with Notes upon the Development of Enamel in general. Comparison of Magnetic Instruments at Kew Observatory. 155 Dr. Monp, Professor Ramsay, and Dr. J. SuHretps. On the Occlusion | of Oxygen and Hydrogen by Platinum Black. Part II. J. EH. Murray. On Contact Electricity of Metals. W. Garpiner. The Histology of the Cell Wall, with Special Refer- ence to the Mode of Connection of Cells. H. L. Cantenpar and H. T. Barnes. On the Variation of the Elec- tromotive Force of Different Forms of the Clark Standard Cell with Temperature and with Strength of Solution. Lord RaytricH. On the Viscosity of Hydrogen as affected by Moisture. The following Papers were read :— I. “ Account of a Comparison of Magnetic Instruments at Kew Observatory.” By C. Curez, Sc.D., F.R.S., Superintendent. II. ‘‘ Note on the Influence of very Low Temperatures on the Germi- native Power of Seeds.” By Horace T. Brown, F.R.S., and F, Escomse, B.Sc., F.L.S. III. “On the Structure and Affinities of Fossil Plants from the Paleozoie Rocks. II. On Spencerites, a new Genus of Lycopodiaceous Cones from the Coal-measures, founded on the Lepidodendron Spenceri of Williamson.” By D. H. Scott, M.A., Ph.D., F.R.S., Hon. Keeper of the Jodrell Laboratory, Royal Gardens, Kew. TV. “ Antagonistic Muscles and Reciprocal Innervation. Fourth Note.” By C.S.SuHerrineron, F.R.S., and Dr. E. H. Herne, At the request of the President, Mr. Gardiner made an oral state- ment on the subject of his Paper published during the vacation (see list above). “ Account of a Comparison of Magnetic Instruments at Kew Observatory.” By C. Cured, Sc.D., F.&.8., Superintendent. Received October 26,—Read November 18, 1897. Last July, M. T. Moureaux, of the Parc Saint-Maur Observatory, near Paris, brought over to England the travelling instruments em- ployed in his magnetic survey of France, and a comparison was made between these and the standard magnetic instruments at Kew Observatory. At the expressed desire of the Kew Observatory Com- mittee, I submit on their behalf a brief account of the comparison and its results. : The comparison serves to connect the standard instruments at Kew Observatory with the standard French instruments at Parc Saint-Maur, the latter, as M. Moureaux has had the goodness to nN 2 156 Dr. C. Chree. inform me, being in excellent agreement with his travelling instru- ments. Parc Saint-Maur may be regarded as the base station for M. Moureaux’s great survey of France and Algeria, while Kew Observatory performed a similar function in the surveys of Great Britain and Ireland, by Professor Riicker and Dr. Thorpe. The existence of the English Channel introduces uncertainty into any conclusions based on the trend of the magnetic lines in France and England, and the instruments employed in the two countries are sufficiently dissimilar to justify scepticism as to their close agreement in the absence of direct experiment. The interest of the comparison is thus far from being limited to the two observatories most imme- diately concerned. ' ; M. Moureaux’s observations at Kew Observatory occupied the afternoon of July 26, and the forenoons of July 27, 28, and 29. On _ the afternoons of the last three days, observations were made with the Kew standard instruments, by Mr. T. W. Baker, chief assistant at the Observatory. All the observations were made in the “ mag- netic house” in the Observatory garden. The comparison was really between M. Moureaux’s absolute instruments and the Kew absolute instruments, but the observations, being made at different hours of the day, had to be connected through the intermediary of the curves from the self-recording magnetic instruments. The elements recorded photographically are the decli- nation, horizontal force, and vertical force. The value in magnetic units of 1 cm. of the ordinates is known, but the value of the base lines, ‘answering to zero ordinates, of the several curves is to a certain extent variable. The usual practice at Kew Observatory is to treat each month separately, deducing the value of the base line for any element from a comparison of the absolute observations for that month with the curve ordinates at the times of the observations. In the case of the declination and horizontal force, the standard- ization of the curves is comparatively simple. In the case of the vertical force, the influence of temperature is unfortunately some- what large, a rise of 1° F. equalling in effect a fall of 00001 C.G.S. unit in the vertical component. There is also the complication that what the curve gives is the vertical force, while what the absolute instrument gives is the inclination. Thus to compare inclinometers used at different hours, one has to. follow a circuitous route by way of the horizontal and vertical com- ponents, allowing a correction for changes of temperature in the magnetograph room during the observations. M. Moureaux observed the inclination early, and Mr. Baker late, in the day, and there happened to be a slight difference in the mean temperature of the magnetograph room at the times of their obser- Vations. Comparison of Magnetic Instruments at Kew Observatory. J57 Taking into consideration the above facts, and the further fact that M. Moureaux’s visit occurred at the end of a month, it was decided to standardise the curves exclusively from Mr. Baker’s special observations, on July 27 to 29. These gave three or more complete determinations of each element, under conditions which might be described, on the evidence of the curves, as an almost per- fect magnetic calm. Mr. Baker’s absolute observations and the corresponding curve measurements were in good agreement, especially in the case of the horizontal force, where the individual calculated values for the base line of the curves showed no difference greater than 0:00002 C.G.S. unit. : Owing to the less direct method of comparing the inclinometers, I regard the results obtained for them as somewhat less trustworthy than the others. The figures under the heading “‘ Observatory—Moureaux ” are to be regarded as the excess in the readings of the absolute Kew instru- ments over those of M. Moureaux’s instruments, supposing the former to have been read simultaneously with the latter. The times specified are actually those occupied by M. Moureaux’s obser- vations. Declination. Kew Observatory— Date. Time. Observatory. Moureaux. Moureaux. July 26 3.47— 4.2 P.M. 4? 6:9! EW pees i + 0°2’ 27 10.5 —10.18 a.m. 4°8 5°0 a a apm 7 16.22—-10.32 ,, 6-0 57 +0°3 28 94 2 0.24, 21 13 +0°8 28 9.28— 9.40 ,, 3°0 19 +11 29 ESf——11.49 ';, a9 93 +0°6 Means... +0°5’ Horizontal Force.* Observatory— Kew Moureaux. Observatory. Moureaux. (Unit being 10-° Date. Time. CEs. C.G.8. C.G.S. unit). July 26 4.138— 4.42 P.M. 0°18354: 0°18356 — 2 27 10.41—11.10 a.m. 25 24 eee dL 28 9.48—10.18 _,, 28 49 —21 eo) 1043—11.11 ,, 20 39 —19 29 10.24—10.56 _,, 20 43 —23 Se 17 11.81 ,, 20 27 sss Mean.... —0°00012 C.GS. unit. * [Nov. 15, 1897. M. Moureaux requests me to explain that in the present com- parison of horizontal force at Kew—as well as in his recent comparisons at 158 Dr.-C. Chree. Inclination. Kew Observatory— Date. Time. Observatory. Moureaux. Moureaux. July 27 11.15—11.41 a.m. 67° 20°2’ 67° 18:97 + 1:3’ 28 q2t 1 AS 20'3 1857 +1°6 29 8.55— 9.18 ,, 19°8 17°3 +2°5 29 9.21— 9.43 ,, 20°0 17°6 + 2°4 29 9.47—10.9 __,, 20°0 18°0 +2°0 Mean. ..c.0) 322 In judging of the results several things merit consideration. Neither inclinometer read to Jess than 1!, decimals arising from arithmetical operations. The Kew unifilar magnetometer reads to 10", but with M. Moureaux’s much smaller instrument readings could not be taken to less than 30”. The great skili of the two observers is beyond question, and the mean of several results obtained without mental bias may possess an accuracy greater than that which any individual reading can lay claim to. Still, personally, I should be very sorry to claim accuracy of the order of 1 in the last significant figure of the mean differences. In the use of the results, one should remember the possibility, or rather probability, of the occurrence of change in magnetic instru- ments. This is a vicissitude to. which travelling instruments are probably most exposed, but even in an observatory standard it is certainly not impossible. The constant “P,” appearing in the factor 1—Pr-?, which allows for the departure of the horizontal force magnet from the ideal infinitely short magnet, appears to be to some extent variable, at least in the Kew instrument. This par- ticular variation does not necessarily affect the calculated horizontal force, because the proper value of ““P” for a special set of obser- vations can be calculated from the experiments themselves. This precaution was, in reality, actually adopted in the present case. Still the fact that a change does take place, which, if undetected, would appreciably influence the results, shows that assumptions of absolute constancy are at present acts of faith, not reason. Until there exists a regular system of intercomparison of the instruments at different observatories, our information on this head is likely to be limited. Pavlovsk (St. Petersburg) and Uccle (Brussels)—he has made use of new values for the constants of the French instruments, which it is intended to apply con- sistently after Jan. 1, 1898. This change, entailing a reduction of 0°00067 C.G.S. vnit in the corrected readings of the French standard, was postponed until the French survey should be completed, though the necessity for it has heen recognised for some years. Par ticulars will be found in a paper by M. Moureaux in the ‘Annales du Bureau Central Météorologique de France,’ vol. 1, 1896, now in the press. | Comparison of Magnetic Instruments at Kew Observatory. 159 After this remark, I think I may safely utilise the present com- parison to extend a table,* in which Professor Riicker and Mr. W. Watson embodied the results of their comparison of the standard instruments at various English and Irish observatories, made by means of travelling instruments, in 1895. The differences between the declination, horizontal force, and inclination instruments are given in order, one below the other, the unit in the case of the horizontal force being 0°00001 C.G.S. unit. The table is to be read as follows :—‘‘ Standard declinometer at Kew reads higher than that at Pare Saint-Maur by 0°5', but lower than that at Falmouth by 0°8’,” and so on. The last column gives the differences from the mean instrument, so to speak, of the five observatories. Pare Fal- Stony- Kew. Saint-Maur. mouth. hurst. Valencia. Mean. — + 0°5’ — 08’ = oeen Gal 0:0’ + 02’ ee ee SY aT — 6 +29 ub Vrseres ae ae 2:0’ Be 1:6’ 2°2/ » 3 1:8’ 4: 0:2' : — 0°5’ — — 1°3’ + 0'6’ — 05’ — 03’ ea 42>. = =. + 6 +41 ra a 2 Cy — — 3°6/ + 0°2’ — 3°8’ — 1 °§’ = OS + 1:3’ — + 1°9’ + 0°8’ + 1:0’ Falmouth... < +18 + 6 — +12 +47 +17 + 1°6’ + 3°6/ — + 3°8/ — 02’ + 1:8’ — 1°)’ — 0°5/ = 4-9’ — — ll’ — 0:9’ Stonyhurst.. 4 + — 6 —12 — +35 + 5 — ‘2:2’ — 02’ — 3°38’ — — 40’ a, ' OO 0/005") 08) 11 — + 02! \eane ng —29 Ad —47 —35 et —30 } + 1:8’ + 3'8’ +. .0°2/ + 4:0’ — 0) The apparent agreement between the standard instruments at Pare Saint-Maur and Stonyhurst is noteworthy. The fact that the Kew standard instruments agree so closely with the imaginary mean instruments was, it may be observed, not noticed by the writer until after he had constructed the table. In his opinion the phenomenon is probably purely fortuitous. In searching, however, for explana-— tions of the discrepancies between the several instruments, or in attempting to remove them, a consideration of the devartnres from the means might be profitable. * ‘Brit. Assoc. Report’ for 1896, p. 97. 160 Messrs. H. T. Brown and F. Escombe. ‘Note on the Influence of very Low Temperatures on the _ Germinative Power of Seeds.”- By Horace T. Brown, F.R.S., and F. Escomse, B.Sc., F.L.S. Received September 27,—Read November 18, 1897. A considerable difference of opinion still exists amongst biologists as to the condition of the protoplasts of resting seeds, spores, &c., in which all ordinary signs of life may be unrecognisable for a consider- able period. According to one view, the essential elements of the cell, during the period of inertness, are still undergoing feeble but imperceptible alteration, accompanied by gaseous exchange with the surrounding atmosphere; and, even when ordinary respiration is in abeyance, it is assumed there are small internal changes going on, due to the inter- action of certain constituents of the protoplasm, reactions which may be independent of the outside gaseous medium, and which are often referred to under the somewhat vague term of “intramolecular respiration.” On the other hand, Pe is sometimes maintained that all metabolism is completely arrested in protoplasm when in the dormant state, and that it then loses, for the time being, all power of internal adjust- ment to external conditions. According to one view, therefore, the machinery of the dormant protoplasts is merely “slowed down” to an indefinite extent, whilst according to the other it is completely brought to rest for a time, to be once more set going when external conditions are favourable. It appears to us that the advocates of the “ slowing-down” hypo- thesis have scarcely given sufficient attention to the experimental evidence available, and that they have been somewhat biassed by a Supposed analogy between the dormant state of seeds and the hiber- nating state in animals, and have also, perbaps, been unconsciously influenced by Mr. Herbert Spencer’s well-known definition of life, which implies a constant internal adjustment in the living protoplasm. The experiments of the late G. J. Romanes, which were described in a short paper laid before the Society in 1893* are full of interest in their bearing on this question. Seeds of various plants were sub- mitted in glass tubes to high vacua of 1/1000000 of an atmosphere for a period of fifteen months. In some cases, after the seeds had been am vacuo for three months, they were transferred to other tubes charged respectively with oxygen, hydrogen, nitrogen, carbon monoxide, carbon dioxide, hydrogen sulphide, aqueous vapour, and the vapour of ether and chloroform. The results proved that neither a high vacuum, nor subsequent exposure for twelve months to any of * ‘Roy. Soc. Proc.,’ vol. 57, p. 335. Injluence of very Low Temperatures on Germinative Power. 161 the above gases or vapours, exercised much, if any, effect on the sub- sequent germinative power of the seeds employed. These experiments of Romanes are certainly of the highest im- portance, since the seeds were submitted for a long period to condi- tions which must certainly have excluded anything like respiration by ordinary gaseous exchange, but the conditions did not preclude with the same certainty the possibility of chemical interactions of some kind or other within the protoplasm, those ill-understood changes, in fact, which have been referred to “intramolecular respiration.” It is true that in some of the experiments, notably those in which the vapours of chloroform and ether were employed, the probability of | any such internal reactions is rendered somewhat remote, but still, in most cases, the experiments admit of the possibility of feeble metabolic activity continuing in the cytoplasm. It occurred to us, some months ago, that more evidence would probably be forthcoming on these points if we could submit seeds to a temperature below that at which ordinary chemical reactions take place, thus eliminating any -possibility of interactions between the constituents of the protoplasm. Owing to the kindness of Professor Dewar, who has been good enough to place the resources of his laboratory at our disposal, and to undertake this part of the work for us, we have been able to ascertain how far the subsequent germinative power of a consider- able variety of seeds is affected by prolonged exposure to the very low temperatures produced by the slow evaporation of liguid azr. The seeds, enclosed in thin glass tubes, were slowly cooled, and immersed in a vacuum-jacketted flask containing about 2 litres of the liquid air, which was kept replenished so as to submit the seeds for 110 consecutive hours to a temperature of from —18&3° C. to —192° C. About 10 litres of liquid air were required for the experi- ment. The seeds had been Oa air-dried only, so contained from about 10 to 12 per cent. of natural moisture. After the above treat- ment they were slowly and carefully thawed, a process which occupied ~ about 50 hours, and their germinative power was then compared with control experiments made on other portions of the seed which had not been treated in any way. The seeds experimented on were as follows :— Hordeum distichon. Trigonella foeenum-groecum. Avena sativa. Impatiens balsamina. Cucurbita Pepo. Helianthus annuus. Cyclanthera explodens. | Heraclewm villoswm. Lotus Tetragonolobus. | Convolvulus tricolor. Pisum elatius. | Funkia steboldiana. 162 Messrs. H. T. Brown and F. Escombe. These include representatives of the following natural orders :— Graminese, Cucurbitacee, Leguminosee, Geraniacess, Composite, Umbelliferee, Convolvulacez, and Liliacez. Some of the seeds are endospermous, others non-endospermous, and the reserve material consists in some cases of starch, and in others of oil or of mucilage. Their germinative power, after being submitted to the low tem- perature, showed no appreciable difference from that of the controls, and the resulting plants, which were in most cases grown to full maturity, were equally healthy in the two cases. In 1892 Professor Dewar and Professor McKendrick found that a temperature of —182° C. continued for one hour is insufficient to sterilise putrescent substances such as blood, milk, flesh, &c., and that seeds would germinate after the action of a similar temperature for the same period of time.* When we commenced our experiments we were unaware that any other observations of a similar nature had been made, but whilst they were in progress our attention was drawn to an important memoir by C. de Candolle,j in which the latent life of seeds is discussed in the light of a number of low temperature experiments made princi- pally by. himself and R. Pictet, and described at intervals in the Geneva ‘ Archives.’{ In the earlier experiments of C. de Candolle and Pictet, made in 1879, temperatures of —39° C. to —80° C. were employed, and these only from two to six hours, whilst Wart-. mann in 1881 exposed seeds for two hours to —110° C. without effect. In 1884 Pictet found that an exposure of various kinds of Bacte- riacev for three days to —70° C., and afterwards for a further period of thirty-six hours to —120°, did not destroy their vitality, and in the same year Pictet and C. de Candolle exposed seeds to —100° C. for four days with the same result. Pictet, in 1893, further extended his observations to various microbes and also to a large number of seeds, and claims to have cooled them down without effect to nearly —200° C., but he gives no details of the experiments, nor any indication of the length of time during which the cooling lasted. His conclusions, however, are that, since all chemical action is. annihilated at —100° C., life must be a manifestation of natural laws of the same type as gravitation and weight. In his memoir of 1895 (loc. cit.) C. de Candolle discusses very * © Roy. Inst. Proc.,’ 1892, vol. 18, p. 699. + ‘Archives des Sci. Phys. et Nat.,’ Geneva, 1895, vol. 33, p. 497. +t E. Wartman, 1860, ‘Archives des Sci. Phys. et Nat.,’ 1860, p. 277; C. de Candolle and Pictet, 1879, ibid., vol. 2, p. 354; ibid., vol. 2, p. 629; E. Wart- mann, 1881, idid., vol. 5, p. 340; R. Pictet, 1884, idid., vol. 11, p. 820; R. Pictet and C. de Candolle, idid., p. 325; R. Pictet, 1893, ibid., vol. 80, p. 293; C. de- Candolle, 1895, idéd., vol. 33, p. 497. Influence of very Low Temperatures on Germinative Power. 163: fully whether we must regard the life of the resting seed as com- pletely arrested for a time or merely temporarily slackened (ralentie),. and he gives the results of some new experiments on seeds maintained at from —37° C. to —53° C. in the ‘“‘snow-box”’ of a refrigerating: machine for a period of 118 days. Most of the seeds resisted this treatment successfully, and the author concludes that after a suffi- cient interval of time has elapsed the protoplasm of the ripe seed. passes into a state of complete inertness in which it is incapable either of respiration or assimilation, and that whilst in this condition it ean support, without detriment to its subsequent revival, rapid and considerable lowering of temperature. De Candolle then points out that if it really be a fact that the suspended life of a resting seed is in any way dependent on respira- tion, we might expect this to be shown by submitting seeds to a barometric vacuum for some time. He does not appear to have followed out this suggestion, and is apparently unaware of the direct experiments on this point carried out by Romanes two years pre- viously; he argues, however, that if ordinary respiration is a factor of any importance, this may be determined by immersing the seeds. in mercury for such a length of time as to ensure the complete con- sumption of the small amount of oxygen contained within their tissues. It was found that when seeds of wheat, and of Lepidium sativum were thus treated, for periods varying from one to three months, their power of germination was not sensibly affected. Although these last described experiments of C. de Candolle go far to show thatany considerable amount of respiration is unnecessary for the maintenance of potential life in the protoplasm of resting seeds, they are not inconsistent with the view that some minute amount of gaseous exchange may be going on during the whole course of the experiment at the expense of the oxygen contained in the seeds at the time of immersion in the mercury. The results. would have been far more conclusive on this point if it had been shown that the gaseous oxygen originally contained in the seed. tissue had been completely used up in an early stage of the experi- — ment. The experiments of Romanes, however, conducted with high vacua and atmospheres of various gases, leave no room for doubt on this question, and we must consequently abandon all idea of the dormant state of resting seeds having any necessary dependence whatever on ordinary respiratory processes. Neither set of experi- ments, however, excludes the possibility of molecular interchanges. in the protoplasm itself, such molecular transpositions in fact as those which can often be induced in living cells placed under anaérobic conditions, and which are all exothermic in character, and generally,. but not necessarily, attended with the liberation of more or less CQ... The great value cf the low temperature experiments we have described 164 Messrs. H. T. Brown and F. Escombe. lies in the fact that such processes of autoxidation, and in fact any conceivable internal chemical change in the protoplasts, are rendered impossible at temperatures of —180° C. to —190° C., and that we must consequently regard the protoplasm in resting seeds as existing in an absolutely inert state, devoid of any trace of metabolic activity, and yet conserving the potentiality of life. Such a state has been admirably compared by C. de Candolle with that of an explosive mixture, whose components can only react under determinate con- ditions of temperature ; as long as these conditions remain unfulfilled the substances can remain in contact with each other for an indefinite period without combining. It appears to us that the occurrence of a state of complete chemical inertness in protoplasm, without a necessary destruction of its potential activity, must necessitate some modification in the current ideas of the nature of life, for this inert state can scarcely be included in Mr. Herbert Spencer’s well-known definition, which implies a continuous adjustment of internal to external relations.* The defi- nition doubtless holds good for the ordinary kinetic state of proto- plasm, but it is not sufficiently comprehensive to include protoplasm in the static condition in which it undoubtedly exists in resting seeds and spores. The definition becomes in fact one of “ vital activity ”’ rather than of life. As it is inconceivable that the maintenance of “‘ potential vitality ”’ in seeds during the exposure of more than 100 hours to a temperature of —180° to —190° C. can be in any way conditioned by, or correlated with, even the feeblest continuance of metabolic activity, it becomes difficult to see why there should be any time limit to the perfect stability of protoplasm when once it has attained the resting state, provided the low temperature is maintained; in other words an immortality of the individual protoplasts is conceivable, of quite a different kind from that potentiality for unending life which is manifested by the fission of unicellular organisms, and with which Weismann has rendered us familiar. In what manner and to what extent “resting”? protoplasm differs from ordinary protoplasm we do not at present know, but there are indications, notably those afforded by the resting state of desiccated * The following passage from the ‘ First Principles’ (Section 25) clearly shows that the author in constructing his definition had not anticipated the possibility of a living organism attaining a state of absolutely stable equilibrium. “ All vital actions, considered not separately but in their ensemble, have for their final purpose the balancing of certain outer processes by certain inner processes. There are unceasing external forces tending to bring the matter of which organic bodies consist into that state of stable equilibrium displayed by inorganic bodies; there are internal forces by which this tendency is constantly antagonised, and the per- petual changes which constitute life may be regarded as incidental to the mainten- ance of the antagonism.” Influence of very Low Temperatures on Germinative Power. 165: Rotifera, and also by some recent experiments of Van Eyck on dis- continuous germination,* that ordinary protoplasts may, by suitable treatment, be brought to the “ resting” condition. . In 1871, Lord Kelvin, in his Presidential Address to the British Association, threw out the suggestion that the origin of life as we know it may have been extra-terrestrial, and due to the ‘‘ moss-grown fragments from the ruins of another world,” which reached the earth as meteorites.t That such fragments might circulate in the intense cold of space for a perfectly indefinite period without pre- judice to their freight of seeds or spores, is almost certain from the facts we know about the maintenance of life by “resting” proto- plasm; the difficulties in the way of accepting such a hypothesis certainly do not lie in this direction. We must express our thanks to Mr. Thiselton- Dyer and to Dr. D. H. Scott, for the facilities they have given us in carrying out these experiments in the Jodrell Laboratory. Addendum. After the completion of the above Note, our attention was called toa recent investigation by M. R. Chodat, on the influence of low tempera- tures on Mucor mucedo.t He found that a lowering of temperature for: several hours to —70° to —110° C. failed to kill young spores of the mucor, and he adduces certain evidence, which is not, liowever, wholly convincing, that even the mycelium itself, when cultivated on Agar- Agar, and whilst in active growth, is able to resist the action of these low temperatures. The author sums up his opinion as to the bearing of his experiments on the nature of life in the following words :— “La respiration elle-méme est évidemment complétement arrétée a cette température ot. les corps chimiques ne réagissent plus les uns | sur les autres. Silon considére que la vie consiste principalement en un échange continuel de substance, soit par la nutrition intra- cellulaire, soit par la respiration, alors il faut convenir qu’a ces tem- peratures basses la vie n’existe plus. C’est une fatale erreur qu’on > rencontre dans presque tous les traités que la pee aie est une condition nécessaire de la vie, alors qu’elle n’est qu’une des con- ‘ditions de sa manifestation. La vie est conditionnée par certaines. structures. les forces quiles mettent en jeu peuvent étre des forces toutes physiques. Hlles sont simplement les sources d’énergie qui pourront mettre la machine en mouvement.” * “ Ann. Agron.,’ vol. 21 (1895), p. 236. + We find that Professor Dewar called attention in one of his Eoyal Institution lectures in 1892 to the bearing of his low temperature experiments with spores, &e., on this suggestion of Lord Kelvin’s (see ‘ Roy. Inst. Proc.,’ 1892, vol. 13 p- 699). t ‘Bulletin de Herbier Boisier,’ vol. 4 (1896), p. 894. 166 Dr. D. H. Scott. On the Structure and “On the Structure and Affinities of Fossil Plants from the Paleozoic Rocks. II. On Spencerites, a new Genus of Lycopodiaceous Cones from the Coal-measures, founded on the Lepidodendron Spencert of Williamson.” By D. H. Scott, M.A., Ph.D., F.R.S., Hon. Keeper of the Jodrell Laboratory, Royal Gardens, Kew. Received November 2, —Read November 18, 1897. (Abstract. ) The fossils which form the subject of the present paper are ‘Cryptogamic strobili, showing evident Lycopodiaceous affinities, but differing in important points from other fructifications of that family, so that it appears necessary to establish a new genus for their recep- tion. Two species are described, one of which (Spencerites insignis) is already known to us from the investigations of Williamson, who named it first Lepidostrobus insignis, and aiterwards Lepidodendron Spencert,* while the other (Spencerites majusculus) is new. In one of his latest publications, Williamson pointed out that it might ultimately be necessary to make his Lepidodendron Spenceri the type of anew genus.f The separation thus suggested is now carried out, on the basis of a renewed investigation of the structure of this fossil. Spencerites insignis 1s a pedunculate strobilus; the vegetative organs are not as yet identified. The specimens are calcified, and their structure admirably preserved. The anatomy of the axis is of a simple Lycopodiaceous type, but differs in details (such as the course of the leaf-trace bundles) from that of the axis of Lepidostrobus. The peduncle bears sterile bracts, similar to the sporophylls of the cone itself; the latter are arranged spirally, or in some cases in alternating verticils. The individual sporophylls are of peltate form, consisting of a short cylindrical pedicel, expanding into a relatively large lamina. The sporangia are approximately spherical bodies; unlike those of Lepidostrobus, they are quite free from the pedicel, and are attached by a narrow base to the upper surface of the lamina, where it begins to expand. The details of the sporangial wall are quite different from those of Lepidostrobus, and the spores are also characteristic. In size they are intermediate between the microspores and macrospores of * Williamson, ‘‘ Organization of the Fossil Plants of the Coal-measures,” Parts 1X, X, XVI, and XIX, ‘Phil. Trans.,’ 1878, 1880, 1889, and 1898. + General Index, Part II, 1893, p. 24. Affinities of Fossil Plants from the Palwozoic Rocks. 167 Lepidostrobus. They are of tetrahedral form, becoming spheroidal when mature, and each spore has a hollow, annular wing running round its equator. The wing is no doubt formed by a dilation of the cuticle,* and not, as Williamson supposed, from the abortive sister- cells. Spencerites majusculus, the new species, is much larger than the former, the axis of the cone being twice as thick. The anatomy is similar, but the sporophylls, and consequently the leaf-traces, are more numerous. The sporopbylls, which are arranged in alter- nating verticils, are relatively short, and of peculiar form ; the lamina is very thick, and of great tangential width. The sporangia are like those of the former species, and similarly inserted, but the spores are quite different. They are smaller than those of S. insignis, and have the form of quadrants of a sphere, with narrow wings along their three angles. The genus is separated from Lepidostrobus, mainly on account of the very different mode of insertion of the sporangia, a character which is accompanied by differences in the form of the sporophylls and sporangia, the structure of the sporangial wall and of the spores, and the whole habit of the strobilus. : Spencerites, and especially S. insignis, bears a considerable re- semblance to the Sigillardostrobus Crepini, of Zeiller, but cannot be united with the genus Sigillariostrobus, for the insertion of the sporangia in the latter, as shown in the Sigillariostrobus ciliatus of Kidston, is totally different. The author is much indebted both to M. Zeiller and Mr. Kidston, for the loan of their specimens for examination. The generic and specific characters may provisionally run as follows :— Spencerttes, gen. nov. Cone consisting of a cylindrical axis, bearing numerous simple sporophylls, arranged spirally, or in crowded alternating verticils. Sporophylls short, formed of a sub-cylindrical pedicel, expanding into a large peltate lamina. Sporangia solitary on each sporophyll, inserted, by a narrow base, on the upper surface of the lamina, but free from the pedicel. Sporangial wall consisting of a single layer of prosenchymatous cells. Spores winged. 1. Spencerites insignis, (Will). Lepidostrobus, sp., Will. “Organization of the Fossil Plants of the Coal-measures,” Part 9, ‘ Phil. Trans.,’ 1878, p. 340, figs. 39 to 47 and 52 to 57. 3 * Cf. Solms-Laubach, ‘ Fossil Botany,’ p. 239. 168 Structure and A finities of Fossil Plants. Lepidostrobus imsignis, Will. Loc. cit. Part 10, ‘ Phil. Trans.,” 1880, p. 502, figs. 11 and 12. Lepidodendron Spencert, Will. Loc. cit. Part 16, ‘ Phil. Trans.,’ 1889, p. 199, figs. 19 to 22; \Part 19, ‘Phil. Trane eigen o4, figs. 41 to 50. Cone pedunculate; peduncle bractigerous. Whole cone 8—10 mm. in diameter. Axis, 35—5 mm. in diameter. Sporophylls, 2— 2°5 mm. long; lamina distinctly peltate, vertically elongated. Sporangia approximately spherical. Spores tetrahedral, becoming spheroidal when free, with a hollow equatorial wing. Maximum diameter of spore, without wing, about 0°14 mm.; with wing, about. 0°28 mm. Wood of axis without prominent angles, with or without pith. Outer cortex containing distinct bands of sclerenchyma. Locality, near Halifax and Huddersfield. Horizon, Lower Coal-measures. 2. Spencerites majusculus, sp. nov.—Whole cone about 15 mm. in diameter, axis about 9 mm. in diameter. Sporophylls about 3 mm. long; lamina obscurely peltate, as seen in radial section, but greatly elongated tangentially, attaining a breadth of 3 mm. Sporangia approximately spherical. Spores having the form of quadrants of a sphere, with three narrow wings. Maximum diameter of spore, without wings, about 0°11 mm.; with wings, about 0°15 mm. Wood of axis with about 30, somewhat prominent, angles; with- out pith. Outer cortex uniformly sclerotic. Locality, near Halifax. Horizon, Lower Coal-measures. Proceedings and List of Papers read. 169 November 25, 1897. The LORD LISTER, F.R.C.S., D.C.L., President, in the. Chair. A List of the Presents received was laid on the table, and thanks ordered for them. In pursuance of the Statutes, notice of the ensuing Anniversary Meeting was given from the Chair, and the list of Officers and Council nominated for election was read as follows :-— President.—Lord Lister, F.R.C.S., D.C.L. Treasurer.—Sir John Evans, K.C.B., D.C.L., LU.D. g : Professor Michael Foster, M.A., M.D., D.C.L., LL.D SAAT. —) professor Arthur William Riicker, Ni, A D.Se. Foreign Secretary.—Sir Edward Frankland, K.C.B., D.C.U., LL.D. Other Members of the Council.—Professor William Grylls Adams, M.A.; Professor Thomas Clifford Allbutt, M.D.; Sir Robert Stawell Ball, M.A.; Rev. Thomas George Bonney, D.Sc.; Professor John Cleland, M.D.; Professor Robert Bellamy Clifton, M.A.; Pro- fessor James Alfred Ewing, M.A.; Alfred Bray Kempe, M.A.; John Newport Langley, D.Sc.; Joseph Larmor, D.Sc.; Professor Nevil Story Maskelyne, M.A.; Professor Raphael Meidola, F.C.8.; Pro- fessor Edward Bagnall Poulton, M.A.; William James Russell, Ph.D.; Dukinfield Henry Scott, M.A.; Professor Walter Frank Raphael Weldon, M.A. Professor J. H. van’t Hoff, Professor Henri de Lacaze-Duthiers, Protessor Wilhelm Pfeffer, and Professor Ferdinand Zirkel were balloted for and elected Foreign Members of the Society. The following Papers were read :— I. “On the Geometrical Treatment of the ‘Normal Curve’ of Statistics, with especial Reference to Correlation and to the heory of Mirror.” By W. F. Suepparp, M.A., LLM., formerly Fellow of Trinity College, Cambridge. Communi- cated by Professor A. R. Foasytu, F.R.S. VOL. LXII. Oo 170 If. ice ve VIL. Vill. Mr. W. F. Sheppard. On the Geometrical ‘Mathematical Contributions to the Theory of Evolution. _ IV. On the probable Errors of Frequency Constants and on the Influence of Random Selection on Variation and Correlation.”’ By Karu Pearson, M.A., F.R.S., Professor of Mathematics and Mechanics, University College, London, and L. N. G. Fiton, B.A. “On certain Media for the Cultivation of the Bacillus of Tubercle.” By ArtHur Ransome, M.D., F.R.S. . “ Further Note on the Transplantation and Growth of Mam- malian Ova within a Uterine Foster-mother.” By WALTER Hears, M.A., Trinity College, Cambridge. Communicated by Dr. W. H. Gaskets, F.R.S. “Further Observations upon the Comnarative Physiology of the Suprarenal Capsules.” By Swate Vincent, M.B. (Lond.), British Medical Association Research Scholar, Communicated by HE. A. Scnirer, F.R.S. . “Summary of Professor Edgeworth David’s Preliminary Report on the Boring at Funafuti.” By T. G. Bonney, D.Se., LL.D. (Univ. McGill), F.R.S., Professor of Geclogy in University College, London. ‘*On the Determination of the Indices of Refraction of various Substances for the Electric Ray. II. Index of Refraction of Glass.’ By Jacapis CuunperR Bossz, M.A., D.Se., Pro- fessor of Physical Science, Presidency College, Calcutta. Communicated by Lorp Rayuuten, F.R.S. ‘On the Influence of the Thickness of Air-space on Total Reflection of Electric Radiation.” By Jacapis CHUNDER Bost, M.A., D.Sc., Professor of Physical Science, Presi- dency College, Calcutta. Communicated by Lord Ray eicu, BP ORSe ‘On the Geometrical Treatment of the ‘Normal Curve’ of Statistics, with especial Reference to Correlation and to the Theory of Error.” By W. F. SHEPPARD, M.A., LL.M., formerly l’ellow of Trinity College. Cambridge. Commu- nicated by Professor A. R. Forsytu, F.R.S. Received October 9,—Read November 25, 1897. (Abstract. ) The object of the paper is, in the first place, to simplify and extend the treatment of normal correlation as expounded by Francis Galton and Kar] Pearson; and in the second place to obtain general Treatment of the ‘Normal Curve’ of Statistics. 171 formulz in the theory of error, and to apply them to questions which arise in relation to normal distributions and normal correlation. The method is, throughout, elementary, the use of the differential and integral calculus being avoided, though geometrical infinitesimals are to a certain extent employed. Geometrical. The normal curve is defined by the property that the product of the abscissa and the subtangent is constant; thusif MP is an ordinate from the base. and PT the tangent, OM. MT = a?, O being a fixed point in the base. The area bounded by the curve and the base OMT is called a normal figure. The length 2a is the parameter of the figure. The definition of the curve leads at once to its projective properties, and also to the formula for the mean value of 2+! or xt? where aw denotes the distance of an element of area from the central ordinate. If the normal curve or normal figure is rotated about its central ordinate, it generates the normal surface or normal solid. It is proved geometrically that any vertical section of this solid (7.e., any section by a plane perpendicular to its base plane) is a normal figure of the same parameter as the central sections; and, therefore, if the sections of the surface hy any series of parallel vertical planes are projected on any plane of the series, the curves so obtained are ortho- gonal projections of one another with regard to their common base. The converse propositions are also established geometrically. The volume of the solid is obtained in terms of its central ordinate and of the parameter of vertical sections; and thus it is found that the central ordinate of a normal figure of semi-parameter unity and area unity is 1/27. Let = be any closed curve in the base plane. Then it is shown how to construct a curve whose area shall be proportional to the portion of the solid which lies vertically above ~, 1.e., to the volume which would be cut out of the solid by a cylinder having & for its cross-section. Thus, when = is given, this volume can be measured mechanically. Statistical. Let Land M denote the measures of two co-existent attributcs, _ 1, and M, their mean values, a* and 6? the mean squares cf the respectife deviations from the means, and ab cos D the mean product of the deviations from the means. Then the angle D is called the divergence. The solid of frequency of (U—L,)/asin D and (M—M,)/b sin D, the planes of reference being inclined at an angle D to one another (so that the included angle is 180°—D), is called the correlation-sclid. 0 2 172 On the ‘ Normal Curve’ of Statistics. It is shown that, whatever the laws of distribution may be, the correlation-solid of the distributions of Land M is the same as that of the distributions of JL+mM and l'‘L+m’M, where J, m, I’, 1’ are any constants whatever. If L and M are distributed ‘‘ normally,” and the distributions are independent, the correlatiou-solid will be a normal solid. Hence it follows that the distribution of /i+mM is also normal. Galton’s definition of normal correlation is adopted; his ‘“ coefii- cient of correlation’’ being therefore cos D. It is shown that the correlation-solid of two normally correlated distributions is a normal solid, and, therefore, if the distributions of L and of M are normally correlated, the values of /L+mM are normally distributed, and the distributions of /L+mM and of /’L4m’M are normally correlated. The value of D, in a case of norma! correlation, can be obtained without calculating the means, mean squares, and mean product. If we find the medians L, and M,, and form a table of double classifica- tion, thus :— Below Ly. | Above Ij. td iq) o 4 5 ine) Id es eve then D PPE O 8 If we know the proportions of individuals for which L exceeds values X and X’, and the proportions for which M exceeds values Y and Y', we can, for any particular value of D, construct an area representing the proportion of individuals for which L lies between X and X’, and M between Y and Y’. The simplest case is that in which the distributions of LL and of M are classified in the same way, e.g., according to the “decile” method. The proportions of indi- viduals falling into the 100 classes corresponding to a double decile classification are obtained by constructing a certain figure, which is the same whatever the value of D may be, and moving the figure through a distance equal to D/360° of its whole length. The diagram so obtained contains 100 areas, representing the proportions in the 100 classes in question. 3 The definitions of independence and of normal correlation are ex- tended to any number of distributions, and it is shown that if the distributions of L, M, N....are normal, and either independent or correlated, the values of 7L + mM + nN +....are normally distri- buted. Mathematical Contributions to the Theory of Evolution. 173 Theory of Error. Let a community be divided into a number of classes, the propor- tions in the different classes being 2, 2, 23...., so that 2, + 22, + 23 + ..-. =1. Suppose a random selection of n individuals to be made, the numbers drawn from the different classes being n(z + 4), N(Zz, + €), n(z3 + €3)..-. It is proved geometrically, with the aid of the binomial theorem, that the values of the errors ¢), &, €3.... are normally distributed, and that the distributions are normally cor- related. It follows that the values of any expression of the form DAc = Aye; + Ace, + Azez; +.... are normally distributed. The mean square of SAc is shown to be { ZA*z—(ZAz)*?}+n, and the mean pro- duct of Ace and {Be to be {>ABz—ZAz. 2Ba jn. The applica- tions are of two kinds :— (1) The values of the probable errors in the determination of certain quantities are obtained, and, in particular, the probable ervors in the mean, mean square of deviation, mean product of deviations, and divergence. (2) Formule are obtained for testing particular hypotheses; e.g., whether two distributions (of any kind) are independent; whether a distribution is normal; and whether two normal distributions are correlated. “ Mathematical Contributions to the Theory of Evolution. IV. On the probable Errors of Frequency Constants and on the Influence of Random Selection on Variation and Correlation.” By Karu Pearson, M.A., F.RB.S., and L. N. G. Finon, B.A., University College, London.. Re- ceived October 18,—Read November 25, 1897. (Abstract. ) 1. This memoir starts with a general theorem, by which the probable errors made in calculating the constants of any frequency distribution may be determined. It is shown that these probable errors form a correlated system approximately following the normal law of frequency, whatever be the nature of the original frequency dis- tribution, 7.e., whether it be skew or normal. The importance of this result for the theory of evolution is then drawn attention to. It is shown that any selection, whether of size, variation, or correlation, will in general involve a modification not only of the size, but the variation and correlation of the whole complex of correlated organs. The subject of directed selection, of which this random selection is ouly a special case, is reserved for another memoir, nearly completed. 174 Prof. K. Pearson and Mr. L. N. G. Filon. It is shown that if 2. Normal correlation is first dealt with. rX = 0°67449 and n be the number of observations: 1—7,* . Probable error of a coefficient 7,. of correlation = X nN Probable error of 7, for variations with definite values,* Correlation of errors in two standard deviations = 71)”. Correlation of errors in a. standard deviation and a correlation coefficient = 7/2. Probable error of a regression coefficient for two organs where o, and o, are the two standard deviations. Probable error of a regression coefficient for three organs 1—7123* Correlation between the errors in two correlation coefficients, 7.e., Pio and 13 tt TeV 13 (1 — 193? — 113° — Tyo" +2 1 y2VosT 13) = T23— . 2(1—13") (1—7T 2") Correlation between the errors in two correlation coefficients, 71.e., 139 and 134 (13—7 127" 23) (%4—7 23134) + G 14— 734713) (%3—T 127713) 1 == (713 —T 47s) (Tea— Te u4) te ("14 —T 224) (23 24731) 2 (1 = 112°) (1 = 13s”) 3. Skew variation is next dealt with. First the probable errors and correlations of the errors ot the constants of the curve yx i, oe a= Le n(it pti * This value for the ‘‘ array” was erroneously given as that for the absolute value of 7, in ‘Phil. 'T'rans.,’ A, vol. 189, p. 345, but the statement was corrected in ‘ Roy. Svc. Proc.,’ vol. 61, p. 850. * wa ‘, ‘ f e Mathematical Contributions to the Theory of Evolution. 175 are calculated. The formule obtained have been cited and used in a memoir by Karl Pearson and’ Miss Alice Lee, “On the Distribution of Frequency (Variation and Correlation) of the Barometric Height at divers Stations.”* We may note the following results :— Probable error of mean = Ao/ Vn. Probable error of standard deviation o Van. © +s) B where 8 ~ We, and i. Bs, Bs.....are .Bernoulli’s numbers. Probable error of skewness Sz re I = Xr es 2 approximatels Vee /1+3 (Si)? i - Correlation of errors in mean and standard deviation. ? =f Fe a Ook te Correlation of errors in mean and skewness = 0. Correlation of errors in skewness and standard deviation — AL+2.(p-+1)?S} - _ Thus a random selection of size differing in its mean value from the population mean gives.in all probability an alteration in variability, but none in skewness. Increased size means decreased variabiliy. A random selection altering variability alters also both size and skewuess of distribution. Probable error of modal frequency yo se Ny (1+ see approximatel Lan \ 12p]’ u which is shown to be always less than the probable error of the mean. The correlation of errors in the moments and their probable errors: up to the fourth are also worked out. 4, As a limiting case the probable errors of the skewness, the moments up to the fourth, and of the distance between mean and * About to be published in ‘ Phil. Trans,’ 176 Mr. Swale Vincent. Further Observations upon the mode in the case of arandom selection from a norma] distribution are worked oub. 5. Analogous results are next obtained for the skew frequency Curves: Lv Tr By Tita = 1 die Y= i ( +f = Fi x and ame vtan—'G ° 1 2 + lay Jt is shown that in these cases the mean size, variation, modal frequency, and skewness are in general all such that their errors are correlated. Hence any selection of size modifies both the variation _and skewness of the distribution. This is of considerable importance for the theory of evolution, as in most cases of zoological frequency the distribution is of the second, or tangent curve, type. Hence random selection of size tends to modify not only variation but skew- ness of distribution. The results are too long to be cited, and their application to special cases involves somewhat lengthy, but not complex arithmetic, which in practical cases we have found much shortened by the use of the “ Brunsviga ” calculator. ‘“ Further Observations upon the Comparative Physiology of the Suprarenal Capsules.” By Swaue ViINcENT, M.B. (Lond.), British Medical Association Research Scholar.* Communi- cated by K. A. ScHAFER, I’.R.S. Received November 2,— Read November 25, 1897. (From the Physiological Laboratory, University College, London.) In previous communications} I have given experimental evidence in favour of the view that the paired suprarenal bodies and the interrenal gland of Hlasmobranch fishes correspond respectively to the medulla and cortex of the suprarenal capsules of the higher Verte- brata. I have further stated, as the result of numerous experiments, that the medullary portion of the suprarenal appears to be absent in Teleosts, the suprarenal bodies in this order of fishes consisting solely of cortex. Since performing the above series of experiments my attention has been devoted to the general physiological effects of extracts * The expenses involved in this research have been defrayed by a grant from the Government Grant Committee of the Royal Society. + ‘Physiol. Soc. Proc.,’ March 20, 1897; ‘ Roy. Soc. Proc.,’ vol. 61, p: 64, 1897 ‘Anat. Anz.,’ vol. 18, Nos. 1 and.2, 1897. Comparative Physiology of the Suprarenal Capsules. 177 obtained from suprarenal capsules.* The extracts were made sepa- rately from cortex aud medulla, and injected subcutaneously into various mammals. It was noted that the injection of medullary material was invariably fatal if a sufficiently large dose were administered, while the cortical extracts produced no appreciable physiological effects. In the present communication the above views have been corrobo- rated by testing the effects of the two kinds of gland in Elasmo- branchs and of the cortical suprarenals of Teleosts, when extracts of them are injected subcutaneously into small mammals. Naturally only very small quantities of material have been available for this purpose, but the effects upon mice have been quite definite. The suprarenal bodies obtained from six specimens of Gadus morrhua (weighing in a moist state 0-4 gram) were extracted by boiling. The filtered extract was then injected beneath the skin of the back of a mouse. No effects whatever supervened. Again, the paired bodies from seven specimens of Scyllium canicula (weighing when moist 0°3 gram) were similurly extracted, and the filtrate administered to the same mouse (which had remained in perfect health) a few days later. The animal was immediately and powerfully affected. The breathing became very rapid, the hmbs became weak, the temperature lowered, and death ensued after con- vulsions in less than five minutes. The interrenal gland produced no effects when similarly adminis- tered. [A further experiment with material obtained from Raja clavata has been performed. The “axillary hearts” (anterior paired bodies) were removed from three fair-sized specimens, and found to weigh in a moist state 0'2 gram. The interrenal bodies were also removed, and weighed also 0'2 gram. Extracts were then prepared of euch of these, and injected subcutaneously into two separate mice of as nearly as possible the same weight. The mouse which was injected with the extract from the paired suprarenals, was affected in a few minutes. The respirations were very quick at first, afterwards becoming slower and slower. Paralysis quickly came on, first in the hind limbs. All the four limbs were distinctly stiffened before death, which supervened in two hours after injection, The other mouse, injected with extract of interrenal, died about 24 hours after injection.t— November 15. ] These experiments afford further positive evidence of the homology * * Physiol. Soc. Proc.,’ June 12, 1897; ‘ Journ. of Physiol.,’ vel. 22 (Nos. 1 and 2, Sept. 1), 1897. . : -} This result must be attributed to contamination with the paired bodies, and is analogous to the effect one sometimes obtains upon the blood-pressure when interrenal extract is injected intravenously. ' 178 Mr. W. Heape. On the Transplantation and Growth of of the paired bodies of Elasmobranchs with the medulla of the mammalian suprarenal. The direct evidence in favour of the homo- logy of the interrenal with the cortex of the suprarenal is mostly morphological and histological, and I have detailed this elsewhere.* “Further Note on the Transplantation and Growth of Mam- malian Ova within a Uterine Ioster-mother.” By WALTER Heaps, M.A., Trinity College, Cambridge. Communicated by Dr. W. H. GASKELL, F.R.S. Received November 2,— Read November 25, 1897. In 1890 I recorded} an experiment designed to show that it is possible to make use of the uterus of one variety of rabbit as a medium for the growth and complete foetal development of fertilised ova of another variety of rabbit. The experiment was further undertaken in order to determine what effect, if any, a uterine foster-mother would have upon her foster-childven, aud whether or not the presence, during development, of foreign ova in the uterus of a mother would affect offspring of that mother present in the uterus at the same time. In the experl- ment above reterred to, two fertilised ova were obtained from an Angora doe rabbit which had been inseminated thirty-two hours previously by an Angora buck, and they were inserted into the fallopian tube of a Belgian Hare doe, which had been inseminated three hours before by a puck of the same breed as herself. In due course the Belgian Hare doe littered six young, four of which were Belgian Hares, while the other two were Angoras. There was no trace of any cross in any of these young, the four Belgian Hares and the two Angoras were true bred. The experiment seemed to me to show, so far as a single experi- ment could show, that a uterine foster-mother has no power of modifying the breed of her foster-children, and that her uterus during gestation and the nourishment she supplies to the embryo is analogous to a bed of soil with its various nutrient constituents, I had hoped to follow this experiment with others on a larger scale the following year, but was unable to make the attempt until 1893. That year 1 had extraordinary bad luck with my rabbits. I used Angoras and Belgian Hares as before, and out of ten Angora does used, four had no ova in their fallopian tubes after being satis- factorily covered, two had dead ova, and only four produced seg- * ‘Zool. Soc. Trans.,’ vol. 14, Part III, 1897; ‘ Birm. Nat. Hist. and Phil. Soe. Proc., vol. 10, Part 1, 1896; ‘Anat. Anz.,’ vol. 12, Nos. 9 and 10, 1896. + ‘Roy. Soc. Proc.,’ vol. 48. Mammalian Ova within a Uterine Foster-mother. 179 menting ova. But that was not all: I had a stock of fourteen Belgian Hare does, and of the four which were operated on to receive the ova from the four Angora does, three of them died under chloro- form and only one bore young, and she had only one young one, and that one was a Belgian Hare. In 1896 I again attempted the same experiment, using this time Dutch and Belgian Hare rabbits; and again I failed, but from a different cause. The Dutch rabbits produced segmenting ova, and the Belgian Hare does stood the operation perfectly satisfactorily, but they were bad breeders. Two of them had only one young one each, one had two, and one six young ones; they were all undoubt- edly Belgian Hares. These Belgian Hare does I had kept for one, some of them for two years, without allowing them to breed, and I am inclined to think that was the reason why they were not so prolific as usual. I considered also that their disinclination to breed might operate adversely on the foreign ova which were introduced, and so check their development, This year I made five experiments, using again Dutch and Belgian Hare rabbits. The method adopted was the same as that already described. A Dutch doe was covered by a Dutch buck, twenty-four or thirty hours later a Belgian Hare doe was covered by a Belgian Hare buck; the Dutch doe was then killed, and segmenting ova, by this time divided into two or four segments, were taken trom her fallopian tube and placed into the open anterior end of the fallopian tube of the Belgian Hare doe. The operation is a very simple one. The Belgian Hare doe is put under anesthetics and stretched out on her stomach. A longi- tudinal incision, 2 in. long, is then made through the skin at a place 14 to 33 in. from the anterior edge of the pelvis, and on a level with the ventral border of the lumbar muscles. aS x x x xX x xX Fe ES x X xX X x X x xX x, xX X x xX x xX x xX x X x xX x xX X x &X x X x XxX “syuout Zz | ‘SOOM “UOIFBAT}{NO JO poled Soe eX x X X XK X RK MK eye 6) OG. OS 6c 6e «c x OT “99H ———— 'SYOOM Z 6h -v]noour Jo o9ecy dak. PTR D ‘0 oAS ——— Oe Oe “OATY RL -oduay, cee ae eee rere eee ee en ence ne eees seOnaTT eoeoeeeneeeorveec ee ee ee see ee ae a RONALD seen ee eee eee eee en eren se sees erOqqy see ee cess eeeeseeeeees are punolsd ON vee erence ee eee een sete enon ee es OTUTCT eee un ee ee Oe 88 CA ap Sgn La Sak OLA NI | eae e ee ee ee ee ee ee ee ee ee cees s SONTTCT wee ee eer ener eens cere s aavotg AYITCOTL weet e eee tere were ee se re eseeees OMIT pee rere ene e ee eee e reeves enone ODE woe e eee ee ween teens ee seen eeees OAT eevee eee reese ee eeeeeeeeesees OQTT eae were ee ee ee ee ee ents wees ees OTIC eC ec Cm (@ f ee Lo | eee ee oe ee orn ereeeeo ere ee were e eoe 079T([ . ceeeowaee oe ee oe ee ee eee? ITB punois Oin J OUTYRIOD JHOTATA ‘oruvg sere e ee ee ee eee re ee eeer seers ONT see ee eraeereeeeeesess TAROTG AyaTeoH] Seno ek a 01 | mae oe see eh a kaye 8 RHR, OG punoas ena LPs | ‘ouTyeos ‘quedo aod § pus sanodva 9 ‘codud-jjwm SutUITy—? BIpoyl S&T S&T Gl TEL OaT Sil Tél 6TT VEL O&T 661 a Tit O1T 60T LOT sol €0T COL ‘ONT On certain Media for Dr, A, Ransome. 198 x x * x X X X KUKI x xX x xX x x x x x x x x x x x oe Ook ne x KX X xX MAK aR aH x x. x x xX X x X X Se Sete Sa mae |: ‘SIQUOUL g | ‘SyqUOU Z|] ‘SyoOM F “SyAVULoy, eae "SHOOM Z £6 be (73 LG (79 66 66 Lg £6 (<9 ZG bE IES (T3 ZG j AVIL . £6 (79 £6 (a9 OL Sel | ete “UOTY "UOTFVATI[ND JO sporseg -B]Nd0UT jo o4vq EPS ae a atc ae og oe a, mn (0)! HAT (] e@eeoneve pe ee eer ee ee een2eoaee08 ec ee ee 08 eee Ceo ea ‘SHOTJRATY[UD [01JU0Q—' XT e[Q¥, 2208 © ae eees oe oe eee ee 08 2 eaeoeee we eee De dao ee aS eae sie OAT eo ee se oe oe eee ou | eoee “* *"soqn4y OFVIOd j fae 058 “A ebie ie SERS SONATE Te | -*++9u0gded ouryvley eooeaea Pe eevee ee OV euoyded avsy eoeeeeeeeteeeeee 0741 reeeses UNIS POOTE | euojded zesy | sereees Tndas POOTT “SIP y the Cultivation of the Bacillus of Tubercle. roo Tt will be seen that in many of the tubes a free growth was observed as early as the end of the first fortnight. Out of the total number in this series of 37, in thirty-six instances there was free growth on the medium employed, on both kinds of paper, and all kinds of condensed fluid. Eleven of them were grown at a temperature of about 20° ©. In only one instance was there complete failure (vapour from healthy breath). Most of these tubes have been left intact, in order that they may be inspected; but six of them were removed, stained, and examined microscopically, in order to determine whether they were true cultures ; this they proved to be. Two of the cultures, after two months’ growth, were sent away to be inoculated into guinea-pigs, but both they and the original culture were found to be non-viruleut.* Microscopic Hxamination. Nearly all the earlier cultures, in which there appeared to have been any growth, were submitted to microscopical examimation. In all the specimens in which this examination did not show distinct signs of growth the result was put down as “nil,” even though a small number of bacilli might have been found. These few bacilli might have come from the inoculation. It was not difficult to recog- nise the abundant growth of a true cultivation. These examinations, however, gave remarkable results in a large number of the specimens grown upon paper. Many of the bacilli were gigantic 1n size, and a considerable number of them showed distinct branching. Others were knobbed at one end or at both ends, when they looked like miniature “life preservers.” In many of the specimens the culture seemed to have penetrated into the sub- stance of the paper. The bearing of these researches upon the subject of the prophyl- axis against tuberculosis seems to be of some importance. They prove that any one of the various organically charged vapours, whether coming from healthy or from diseased lungs, from the air of cellars, or from comparatively pure ground, forms an excellent culti- vating medium for the bacillus of tubercle when kept away from the disinfecting influence of air and light. This power of promoting its growth is particularly manifest when the supporting substance is common wall-paper, ro it 19 quite apparent when very pure filter-paper is used. It is further proved that, on these substances, the growth of the bacillus may take place at the ordinary temperatures of dwelling- * A further research, with cultures of the bacillus of undoubted virulence, has now been undertaken. 200 Prof. T. G. Bonney. On Professor E. David's rooms; and, hence, that there is no safety against the increase of the organism in ordinary living rooms in which active tuberculous dust is present, and in which the natural disinfectants of the bacillus, fresh air and light, are not present in sufficient amount to destroy their virulence. “Summary of Professor Edgeworth David’s Preliminary Report on the Results of the Boring in the Atoll of Funafuti.” Communicated by Professor T. G. Bonney, F.R.S., Vice- Chairman of the Coral Reef Boring Committee. Received November 25,---Read November 25, 1897. The boring at Funafuti, according to the latest advices, had reached a depth of 643 feet. Professor David’s report is transcribed from notes made during the progress of the work, and gives his first im- pressions of the materials brought up, down to a depth of 557 feet, which had been reached when he quitted the island to return to his duties at Sydney, leaving the work in charge of his assistant. The latest advices informed him that the boring was arrested at 643 feet, but as it was hoped this was only for a time, we are daily expecting to hear yet more gratifying news. His last letters, received during the present week, give a few particulars of the materials pierced between 557 and 643 feet. The werk, Professor David states, often pre- sented most serious difficulties, which would probably have frustrated their efforts, but for the experience gained on the former occasion. The bore hole is situated about half a mile N.H. of the Mission Church, and its height above ‘sea level is about 1 foot above high water mark at spring tides. The diameter is 5 inches down to 68 feet; it is lined with 5-inch tubing down to 118 feet, and 4-inch from surface to 520 feet, so that on September 6 a 4-inch core was being obtained. The following is a general description of the materials pierced :—- For about a yard at the top there was a hard coral breccia. This was followed down to a depth of 40 feet by ‘coral reef rock,” into the composition of which WHeliopora cerulea, with spines of echinids and nullipores, entered largely, the last predominating over the coral at from 15 to 20 feet. From 40 to 200 feet came more or less sandy material, but with a variable quantity of corals. These were scattered through the sand (calcareous and of organic origin; foraminifera, at about 40 feet, making from one-half to two-thirds of the whole) sometimes as fragments (forming occasionally a kind of rubble), but sometimes in the position of growth. Between 120 and 130 feet, and from about 190 to 200 feet, the material Report on the Boring in the Atoll of Funafuti. 201 is described as fairly compact coral rock, so that very probably reefs tm situ, though of no great thickness, were pierced at these depths. The sand appears to be largely derived from coral, but foraminifera occur, sometimes in abundance; so too do nullipores, and here and there spines of echinids. Towards 150 feet signs of change begin to appear in the corals, and these become more conspicuous as the boring approaches its greatest depth. In such case, if I understand rightly, some of the branching corals crumble away and are repre- sented only by casts, while others remain, the surrounding matrix becoming solid, cemented apparently by calcite. Below 202 feet a decided change takes place in the character of the deposit. All above this seems to be largely composed of material derived from corals, with occasional rather brief interludes of true reef, and this mass, measuring, as said above, rather over 200 feet in thickness, may be termed the first or uppermost formation. Below this, down to about 373 feet, sandy material distinctly dominates, which sometimes is almost a calcareous mud. Still even there coral fragments and rubble occasionally appear, and now and then a few isolated corals. Other organisms may be detected, including nullipores, foraminifera, and mollusea; but until this material has been examined microscopically, it would be premature to attempt any precise statement. This mass, in thickness about 170 feet, may be termed the second or middle formation. It is not reef, though obviously produced in the vicinity of a reef. Below 370 feet is the third or lowest zone; in this beds composed of broken coral become frequent, which are intercalated with masses of dead coral, though sandy bands also occur. The character of the material suggests that it has been formed in the immediate vicinity of a reef, which has occasionally grown out laterally, though only for a time, and has built up a layer of true reef, from 2 to 3 feet in thickness, upon a mass of detrital coral. Tn one place the rock is specially noted as “hard,” and hereabouts even the shells of gasteropods have perfshed, only their casts remain- ing. From 526 to 555 feet the bore passed through fairly compact and (in places) very dense and hard “coral limestone” and “ cav- ernous coral rock.” in which dendroid forms were numerous. As regards the part between 557 feet and 643 feet only brief infor- mation is to hand, but Professor David states that it is reported to be chiefly coral limestone, hard and dense, with occasional soft bands of coral sand or coral rubble. Thus the third, or lowest zone, about 270 feet in thickness, corresponds apparently with the first, but it seems to contain larger and more numerous masses of true reef. Professor David has also forwarded with his latest letters a section of the boring and of the exterior form of the island, down to about 730 fathoms: the one drawn from his notebook, the other from Captain Field’s record of soundings. From this I gather the VOL, LXII. Q 202 Annwersary Meeting. following particulars:—The borehole is, roughly speaking, rather over 100 yards from the margin of the ocean, and about 165 yards from that of the lagoon; it is about 240 yards from the spot where a sounding of 10 fathoms was obtained, nearly 400 yards from a 36-fathom sounding, and rather more than a quarter of a mile from one of 130 fathoms. After this the sub- marine slope, for a considerable depth, is not quite so steep. He also states that, at Funafuti, the vigorous growing portion of the reef appeared to be limited to within about 40 feet of the surface. It would be premature, as Professor David remarks, to express an opinion as to the theoretical bearing of these results until the core has been thoroughly studied. But two things seem clear, (1) that true reef has been pierced at depths down to more than 600 feet, and (2) that throughout the whole of the time represented by the mass which has now been tested, coral must have grown in great abundance in some part or other of the locality now represented by Funafuti; for the atoll, it must be remembered, is surrounded by water about 2,000 fathoms deep, what would completely isolate it from any other coralliferous locality. November 30, 1897, Anniversary Meeting. The LORD LISTER, F.R.C.S., D.C.L., President, in the Chair. A full Report of the Anniversary Meeting, with the President’s Address and Report of Council, will be found in the ‘ Year-book’ for 1897-8. The Account of the Appropriation of the Donation Fund and of the Government Grant will also be found in the ‘ Year-book.’ ee ee Ee a Le CT el a ee ee ee aS ee ee Proceedings and List of Papers read. 2038 December 9, 1897. The LORD LISTER, F.R.C.S., D.C.L., President, in the Chair. A List of the Presents received was laid on the table, and thanks ordered for them. The President announced that he had nominated as Vice-Presi- dents for the ensuing year— The Treasurer (Sir John Evans). Professor Cliftcn. Professor Story Maskelyne. Dr. W. J. Russell. The following Papers were read :— I. “On the Densities of Carbonic Oxide, Carbonic Anhydride, and Nitrous Oxide.” By Lorp Rayteteu, F.R.S. II. “On the Application of Harmonic Analysis to the Dynamical Theory of the Tides. Part II. On the General Integration of Laplace’s Dynamical Equations.” By 8. 8. Hoven, M.A., Fellow of St. John’s College, and Isaac Newton Student in the University of Cambridge. Communicated by Professor G. H. Darwin, F.R.S. {ll. “A Note on some further Determinations of the Dielectric Constants of Organic Bodies and Hlectrolytes at very Low Temperatures.” By James Dewar, M.A., LL.D., F.RB.S., Fullerian Professor of Chemistry in the Royal Institution, and J. A. Fremine, M.A., D.Sc., F.R.S., Professor of Hlec- trical Engineering in University College, London. IV. “On Methods of making Magnets independent of Changes of Temperature; and some Hxperiments upon Negative Tem- perature Coefficients in Magnets.” By J. Recinanp Asu- WoRTH. Communicated by ArrHuR Scuoster, F.R.S. VY. “The Electric Conductivity of Nitric Acid.” By V. H. Vutey, M.A., F.R.S., and J. J. Manuey, Daubeny Curator of the Magdalen College Laboratory, Oxford. VI. “On the Calculation of the Coefficient of Mutual Induction of a Circle and a Coaxial Helix, and of the Electromagnetic Force between a Helical Current and a uniform Coaxial Circular Cylindrical Current Sheet.” By J. Viriamu Jones, F.R.S. @) 2 204 Lord Rayleigh. On the Densities of Carbonic VII. “On the Refractivities of Air, Oxygen, Nitrogen, Aron, Hydrogen, and. Helium.” By Professor WinitAmM Ramsay, Ph.D., LL.D., Se.D., F.R.S., and Morris W. Travers, B.Sc. “On the Densities of Carbonic Oxide, Carbonic Anhydride, and Nitrous Oxide.” By Lorp Rayueicn, F.R.S. Received Octcber 12,—Read December 9, 1897. The observations here recorded were carried out by the method and with the apparatus described in a former paper,* to which reference must be made for details. It must suffice to say that the globe containing the gas to be weighed was filled at 0° C., and toa pressure determined by a manometric gauge. This pressure, nearly atmospheric, is slightly variable with temperature on account of the expansion of the mercury and iron involved. The actually observed weights are corrected so as to correspond with a tempera- ture of 15° C. of the gauge, as well as for the errors in the platinum and brass weights employed. In the present, as well as im the former, experiments I have been ably assisted by Mr. George Gordon. ) Carbonic Oxide. This gas was prepared by three methods. In the first method a flask, sealed to the rest of the apparatus, was charged with 80 grams recrystallised ferrocyanide of potassium and 360 c.c. strong sulphuric acid. The generation of gas could be started by the application of heat, and with care it could be checked and finally stopped by the removal of the flame with subsequent application, if necessary, of wet cotton wool to the exterior of the flask. In this way one charge could be utilised with great advantage for several fillings. On leaving the flask the gas was passed through a bubbler centaining potash solution (convenient as allowing the rate of production to be more easily estimated) and thence through tubes charged with frage- ments of potash and phosphoric anhydride, all-connected by sealing. When possible, the weight of the globe full was compared with the mean of the preceding and following weights empty. Four experi- ments were made with results agreeing to within a few tenths of a milligram. In the second set of experiments the flask was charged with 100 grams of oxalic acid and 500 c.c. strong sulphurie acid. To absorb the large quantity of CO, simultaneously evolved a plentiful * “On the Densities of the Principal Gases,” ‘ Roy. Soc. Proc.,’ vol. 58, p. 134, 18938. r Oxide, Carbonic Anhydride, and Nitrous Oxide. 205 supply of alkali was required. A wash-bottle and a long nearly horizontal tube contained strong alkaline solution, and these were followed by the tubes containing solid potash and phosphoric anhyd- ride as before. For the experiments of the third set owalic acid was replaced by formic, which is more convenient as not entailing the absorption of large volumes of CO. In this case the charge consisted of 50 grams formate of soda, 300 c.c. strong sulphuric acid, and 150 c.c. distilled water. The water is necessary in order to prevent action in the cold, and the amount requires to be somewhat carefully adjusted. As purifiers, the long horizontal bubbler was retained and the tubes charged with solid potash and phosphoric anhydride. In this set there were four concordant experiments. The immediate results stand thus :— Carbonic Oxide. Brom ferrocyamide. ....-.-.-+. 2°29843 SU Oma 1C ACI kia: oo dix ate eels 2°29852 be ontormabe of SOA: secsw'sir' +s 2°29854 : MEGA Min aren hig 0 2°29850 This corresponds to the number 2°62704 for oxygen,* and. is subject to a correction (additive) of 0°00056 for the diminution of the external volume of the globe when exhausted. The ratio of the densities of carbonic oxide and oxygen is thus 2°29906 : 2°62760; so that if the density of oxygen be taken as 32, that of carbonic oxide will be 27:°9989. If, as some preliminary experiments by Dr. Scott+ indicate, equal volumes may be taken as accurately representative of CO and of O,, the atomic weight of carbon will be 11:9989 on the scale of oxygen = 16. The very close agreement between the weights of carbonic oxide prepared in three different ways is some guarantee against the presence of an impurity of widely differing density. On the other hand, some careful experiments led Mr. T. W. Richardst to the conclusion that carbonic oxide is liable to contain considerable quan- tities of hydrogen or of hydrocarbons. From 54 litres of carbonic oxide passed over hot cupric oxide he collected no less than 25 milli- grams of water, and the evidence appeared to prove that the hydrogen was really derived from the carbonic oxide. Such a proportion of hydrogen would entail a deficiency in the weight ‘of the globe of about 11 milligrams, and seems improbable in view of the good agreement of the numbers recorded. The presence of so much * “On the Densities of the Principal Gases,” ‘Roy. Soc. Proc.,’ vol. 53, p. 144. t+ ‘Camb. Phil. Proc.,’ vol. 9, p. 144, 1896. { ‘Amer. Acad. Proc.,’ vol. 18, p. 279, 1891. 206 Lord Rayleigh. On the Densities of Carbonic hydrogen in carbonic oxide is also difficult to reconcile with the well- known experiments of Professor Dixon, who found that prolonged treatment with phosphoric anhydride was required in order to render the mixture of carbonic oxide and oxygen inexplosive. In the presence of relatively large quantities of free hydrogen (or hydro- carbons) why should traces of water vapour be so important ? In an experiment by Dr. Scott,* 4 litres of carbon monoxide gave only 1:3 milligrams to the drying tube after oxidation. I have myself made several trials of the same sort with gas pre- pared from formate of soda exactly as for weighing. The results were not so concordant as I had hoped,f but the amount of water collected was even less than that given by Dr. Scott. Indeed, I do not regard as proved the presence of hydrogen at all in the gas that I have employed.{ Carbonic Anhydride. This gas was prepared from hydrochloric acid and marble, and alter passing a bubbler charged with a solution of carbonate of soda, was dried by phosphoric anhydride. Previous to use, the acid was caused to boil for some time by the passage of hydrochloric acid vapour from a flask containing another charge of the acid. Ina second set of experiments the marble was replaced by a solution of carbonate of soda. There is no appreciable difference between the results obtained in the two ways; and the mean, corrected for the errors of weights and for the shrinkage of the globe when exhausted, is 3°6349, corresponding to 2°6276 for oxygen. ‘The temperature at which the globe was charged was 0° C., and the actual pressure that of the manometric gauge at about 20°, reduction being made to 15° by the use of Boyle’s law. From the former paper it appears that the actual height of the mercury column at 15° is 762°511 mm. Nitrous Oxide. In preliminary experiments the gas was prepared in the labora- tory, at as low a temperature as possible, from nitrate of ammonia, or was drawn from the iron bottles in which it is commercially sup- plied. The purification was by passage over potash and phosphoric anhydride. Unless special precautions are taken the gas so obtained is ten or more milligrams too light, presumably from admixture with * ‘Chem. Soe. Trans.,’ 1897, p. 564, + One obstacle was the difficulty of re-oxidising the copper reduced by carbonic oxide. JI have never encountered this difficulty after reduction by hydrogen. £ In Mr. Richards’ work the gas in an imperfectly dried condition was treated with hot platinum black. Is it possible that the hydrogen was introduced at this stage P Oxide, Carbonic Anhydride, and Nitrous Oxide. 207 nitrogen. In the case of the commercial supply, a better result is obtained by placing the bottles in an inverted position so as to oa from the liquid rather than from the gaseous portion. Higher and more consistent results were arrived at from gas which had been specially treated. In consequence of the high relative solubility of nitrous oxide in water, the gas held in solution after prolonged agitation of the liquid with impure gas from any supply, will contain a much diminished proportion of nitrogen. To carry out this method on the scale required, a large (11-litre) flask was mounted on an apparatus in connection with the lathe so that it could be vigorously shaken. After the dissolved air had been suffi- ciently expelled by preliminary passage of N,O, the water was cooled to near 0° C, and violently shaken for a considerable time while the gas was passing in large excess. The nitrous oxide thus purified was expelled from solution by heat, and was used to fill the globe in the usual manner. For comparison with the results so obtained, gas purified in another manner was also examined. A small iron bottle, fully charged with the commercial material, was cooled in salt and ice and allowed somewhat suddenly to blow off half its contents. The residue drawn from the bottle in one or other position was employed for the weighings. Nitrous Oxide (1896). een Mrepelled from Water: ....600-+6as0rcineee 3°6359 Le Be eae iatiat al anh, oa cm, '0),4-6 48 cee oO Oo04 ,» 19 From residue after blow off, valve downwards 3°6364 cama. ae bs yalve upwards.. 3°6358 a ae Bs 33 valve downwards 3°6360 «GS ind 5 pees eee ere aia oi dra 3°6359 The mean value may be taken to represent the corrected weight of the gas which fills the globe at 0° C. and at the pressure of the gauge (at 15°), corresponding to 2°6276 for oxygen. One of the objects which I had in view in determining the density of nitrous oxide was to obtain, if it were possible, evidence as to the atomic weight of nitrogen. It may be remembered that observations upon the density of pure nitrogen, as distinguished from the atmo- spheric mixture containing argon which, until recently, had been confounded with pure nitrogen, led* to the conclusion that the densi- ties of oxygen and nitrogen were as 16: 14003, thus suggesting that the atomic weight of nitrogen might really be 14 in place of 1405, 'as generally received. ‘The chemical evidence upon which the latter number rests is very indirect, and it appeared that a direct compari- * Rayleigh and Ramsay, ‘ Phil. Trans.,’ vol. 186, p. 190, 1895. 208 On the Densities of Carbonic Oxide, &c. son of the weight of nitrous oxide and of its contained nitrogen might be of value. A suitable vessel would be filled; under known conditions, with the nitrous oxide, which would then be submitted to the action of a spiral of copper or iron wire rendered incandescent by an electric current. When all the oxygen was removed, the residual nitrogen would be measured, from which the ratio of equi- valents could readily be deduced. The fact that the residual nitrogen would possess nearly the same volume as the nitrous oxide from which it was derived would present certain experimental advantages. If indeed the atomic weights were really as 14 : 16, the ratio (a) of volumes, after and before operations, would be given by 2:2996x2 __ 14 3°€359—42996xa2 8’ 073-6350 ee whence ® = 719-9008 ane 3°6359 and 2°2996 being the relative weights of nitrous oxide and of nitrogen which (at 0° C. and at the pressure of the gauge) occupy the same volume. The integral numbers for the atomic weights would thus correspond to an expansion, after chemical reduction, of about one-half per cent. But in practical operation the method lost most of its apparent simplicity. It was found that copper became unmanageable at a temperature sufficiently high for the purpose, and recourse was had to iren. Coils of iron suitably prepared and supported could be adequately heated by the current from a dynamo without twisting hopelessly out of shape; but the use of iron leads to fresh difficulties. The emission of carbonic oxide from the iron heated in vacuum continues for a very long time, and the attempt to get rid of this gas by preliminary treatment had to be abandoned. By final addition of a small quantity of oxygen (obtained by heating some permanganate of potash sealed up in one of the leading tubes) the CO could be oxidised to CO,, and thus, along with any H,O, be absorbed by a lump of potash placed beforehand in the working vessel. To get rid of superfluous oxygen, a coil of incandescent copper had then to be invoked, and thus the apparatus became rather complicated. It is believed that the difficulties thus far mentioned were over- come, but nevertheless a satisfactory concordance in the final num- bers was not attained. In the present position of the question no results are of value which do not discriminate with certainty between 14°05 and 14°00. The obstacle appeared to lie in a tendency of the nitrogen to pass to higher degrees of oxidation. On more than one occasion mercury (which formed the movable boundary of an overflow chamber) was observed to be attacked. Under these circumstances Application of Harmonic Analysis to Tidal Theory. 209 I do not think it worth while to enter into further detail regarding the experiments in question. The following summary gives the densities of the various gases relatively to air, all obtained by the same apparatus.* The last figure is of little significance. @uvutree trom H,O and CO, ........... 100000 oO PH ee ere 1710535 Nitrogen and argon (atmospheric) .... 0:97209 mA ET CM 8S ie SE Sek lela os walle 0°96737 Rema Saratig: ate gels ll Gd walk Kasei & aL (52 Carbonic oxide..... ia ald Ce SP Mates el 0°96716 menbanie anhydride .....0..8.02.0..5- 152909 PEN HOIL CHEN i. gL TIuiC.S saieteiies Sleiel aod obtae - 152951 The value obtained for hydrogen upon the same scale was 0:06960 ; but the researches of M. Leduc and of Professor Morley appear to show that this number isa litle too high, “On the Application of Harmonic Analysis to the Dynamical Theory of the Tides. Part II. On the general Integration of Laplace’s Dynamical Equations.” By 8. 8. douex, M.A., Fellow of St. John’s College, and Isaac Newton Student in the University of Cambridge. Communicated by Professor G. H. Darwin, F.R.S. Received October 27, —Read December 9, 1897. (Abstract. ) The former part of this paper deals with solutions of Laplace’s differential equations for the tides symmetrical with respect to the axis of rotation. In the present part the restriction of symmetry is no longer imposed, and a general solution is sought, the law of depth of the ocean, however, being limited to the case which will admit of both the interior and exterior surfaces being regarded as spheroids of revolution. It is found that, subject to this limitation, if the © solution sought represent a simple harmonic motion of any period whatsoever, and the height of the surface-waves be expressible as an infinite series of tesseral harmonics of the same rank but different orders, a linear relation connecting three successive coefficients of the series can be deduced similar to that obtained in Part I.+ From this relation a period-equation for the free vibrations is deduced, and a method of determining approximate values of the * “Roy. Soc. Proc.,’ voi. 53, p. 148, 1898; vol. 55, p. 340, 1894; ‘ Phil. Trans.,’ vol. 186, p. 189, 1895 ; ‘ Roy. Soc. Proc.,’ vol. 59, p. 201, 1896. + ‘Roy. Soc. Proc.,’ vol. 61, p. 236. 210 Mr. J. R. Ashworth. Methods of making Magnets higher roots is given. The earlier roots are examined numerically, and tabulated for four different depths of the ocean for the types involving tesseral harmonics of rank 1 and 2, these types presenting special interest in connection with the diurnal and semi-diurna! forced tides respectively. The types of free oscillation are found to be of two classes, dis- tinguishable by their limiting forms when the rotation-period is indefinitely prolonged. In the former class the motion remains oscil- latory when the period of rotation becomes infinitely long, while in the latter the “speed” of the oscillation always bears a finite ratio to the angular velocity of rotation, so that the oscillation will be replaced by a steady motion when the angular velocity of rotation is reduced to zero. In dealing with the forced oscillations, the theorem of Laplace that in an ocean of uniform depth there will be no diurnal rise and fall at the surface is obtained and generalised as follows:—In an ocean of uniform depth the tides due to a disturbing potential of degree s+1 and rank s will involve no rise and fall at the surface if the period of the disturbing force be $(s+1) sidereal days. A theorem given by Professor Darwin with reference to the expression of the semi-diurnal tides in finite terms, as also Laplace’s solution of the problem of the diurnal tides in an ocean of variable depth, is found to admit of similar generalisation. The general problem of the forced tides due to any disturbing force derivable from a potential function in the cases where infinite series are required for the solution is treated analytically, and is further illustrated by numerical examples typical of the leadmeg tidal constituents which occur on the earth, the results where possible being compared with those obtained by other methods. “On Methods of making Magnets independent of Changes of Temperature; and some Experiments upon Abnormal] or Negative Temperature Coefficients in Magnets.” By J. REGINALD ASHWORTH, B.Sc. Communicated by ARTHUR SCHUSTER, F.R.S. Received October 29,—Read December SA bre The present investigation, which has been carried out in the Physical Laboratory of the Owens College, Manchester, was under- taken at Professor Schuster’s suggestion with the object of ascertain- ing what kinds of iron and steel are least lable to a change of magnetic intensity under moderate fluctuations of temperature. Specimens of steels containing severally tungsten, manganese, cobalt, and nickel, also cast irons, of different blends of pig irons, and independent of Changes of Temperature. aa ot different percentages of carbon, were procured from a number of different English and Scotch firms. The size of these specimens was in general about 15 cm. long and 1 or 2 thick, but as it was not uniform the dimensions and weight in grams of each are given in the accompanying table, columns I, II, and III. In column IV has been entered the dimension ratio, 7.e., the ratio of the length to the diameter or breadth, so that a comparison may more consistently be made of the magnetic behaviour of any two specimens. For thin rods, Cancani* finds that increase of this ratio tends to diminish the temperature coefficient of a magnet. The course of an experiment was as follows:—The rod or bar in its normal state, or after being hardened or annealed as occasion required, was magnetised between the poles of a powerful electro- maenet excited by a battery.of twenty-six storage cells. The magnet was then fixed rigidly in a horizontal tube, through which a stream of cold water and steam could be alternately passed. The tube and its contents were placed at a convenient distance from a sensitive dead- beat magnetometer and at right angles to the magnetic meridian. The deflections of the magnetometer needle were read by the usual mirror and scale, the distance of the scale from the mirror being 1 metre, and from the readings were deduced directly the tempera- ture coefficient and the total irreversible loss of magnetism. As the deflections were never more than a few degrees of arc the angles an‘ their tangents were virtually equivalent. The intensity of magneti- sation in C.G.S. units or magnetic moment per unit volume, although not necessarily required, was approximately determined from the formula 4 A l(@—#/? tan 6 2 : 2d m in which the earth’s horizontal force, H, was considered throughout as constant and equal to 0°18 C.G.S. unit and also o, the density, was uniformly taken to be 7°8; m signifies the mass in grams; d the distance from the centre of the magnet to the magnetometer needle ; L the half length of the magnet, and 0 the deflection. The process of heating and cooling the magnet was continued until the intensity fluctuated between two nearly constant values corre- sponding to the temperatures of the cold water and steam. The coefficient « given in the eighth column was then calculated by inserting these values in the equation la = (1-2 i’ —t).+ * R. Cancani, ‘Atti della R. Acc. dei Lincei,’ (4), 3, pp. 501—506, 1887 ; ‘ Beibl.,’ vol. 11, 1887. + I have followed the customary mode of writing this formula with a negative sign preceding the coefficient, a; and, hence, a negative coefficient indicates an increase of magnetic intensity with increase of temperature. 212 Mr. J. R. Ashworth. Methods of making Magnets 7 The irreversible loss of original magnetic intensity which results from a series of heatings and coolings is tabulated under the heading 8 in column VII, 6 being calculated from the formula = I; aes where I; and I; are the final and initial intensities. The limits of temperature ¢ and ¢’ in these experiments were 10° to 20° C. and about 100° C., giving a range of 80° or 90°. . The centigrade scale of temperatures and the C.G.S. system of units are to be understood throughout. In every case a record has been kept of the scale readings at the temperatures ¢ and ¢’ during the progress of the operations of heating and cooling, and the brief example here cited may be taken as typical. Three per cent. Tungsten Steel. Temperature. Scale nenglnes at a a =a b. A Zero 0°0 | | 6°5 184°8 | 99°6 ae 142 °6 70 Hal <3 99°6 a 140 °2 7-5 160 °2 99°6 Se 137 °9 75 158°8 99°6 ie 137 °9 fi) 158°6 Zero —O°l The table which is annexed gives a synopsis of the results obtained. Each number in the first column represents a separate piece of iron or steel, but where comparative tests of the same material were desired, as, for example, in the cast irons, when either annealed or hardened, the precaution was taken to employ two pieces of origin- ally a single rod. Thus Nos. 15 and 16 are two parts of the same rod cut through the middle, and similarly with Nos. 17 and 18 and others. In the first place, several varieties of steels were tested. No. 1 is a steel from Sheffield, supplied specially for making magnets; Nos. 2 and 38 are Hadfield’s well-known non-magnetic steel; the next four are tungsten steels, of which Nos. 6and 7 are known as Mushet’s self-hardening steel, having the property of hardening even when cooled slowly. These were both cut from the same rod; No. 6 was 213 ‘qa00 rod @ TaA0 UOGreDA ‘suoat dvaos pure ‘ozrtpas yoqoos | ‘suo. dvaos pure ‘youre su0 [5 4 ‘suoa devaios pure ‘ory sy10 X ‘wo aurourecr 4 ‘og ‘od 2Z.9 moquyQ ‘od z= wt ‘og «od gz.oqnoqgeg ‘o'dg = IN ‘yojoog «od 6TO=9 0 'dp.Z= Ne od ‘o'dar.o=9 ‘0 'dg=IN ‘0d Odeon=o9 ‘0'dap=o0p OT gual ‘og ‘10048 SUTMOpABY-JIOS 8 JoysnL ‘ppwoys mor od ¢.g=1g ‘o'dg.9 = 49 ‘odgqt=uq odtg=9 odtg= mM independent of Changes of Temperature. “PP HECS mort. o'dOT 0 F0=— 9. 9 'dge= A xeYG | ‘Ocf “plPyeyg aoc ee PSO + 660 + $S0 + 910+ CVG + 910 + GLE + 810+ 88g + LO0 ¥20'— slo— V6O + 16) 0 fas GE0 + 00¢ + G20 + Sj] ae L60 + 690 + G60 + GPT + Sto + TE0 + LET + ‘00-0 (2) abi TA ‘d TIA “W0TJD9S SSO.) 9-7 0.¢ 6-16 I- 1& 6-96 6 LG 8.-DZ1| 9-O6L 9: SE 6-9 T-O8T;} 6-016 G+ LE g. 6g 8-291} T-&6T 9-16 T- 6S cV-0 | PS-0 cy-O | %S-0 6-S¢ | 4. TTT T-99 | ¥ 90T L. G& 8-62 F601; O- 8eL 9-61 4-91 L-¥6 L. 86 8-CL | ¥-9T 8: Z4T| §- SIZ 9-GET| 6-C8T 6-09 | 8-69 T- PE L. 68 LS6-0 0€- 0 69-0 | $4-0 9-0F | 6&9 JT 7 | ak L298 li NMA MM MAID MOH ODHHHOOMWH SS eB Oe ee Be | ° > n - n ra rt D COMWGW AK MSOOD Sia > MOOD 6 ‘PllZ=0 gai: G10 | SS-Z|9-ST = 6S | 96-0/0-21| peuepaeyy co | 96-0]8- sT{portddns sy ZIL| L1I-1/%-ST| pouspaeyy SIL| Z1-T{s- sT/porddns sy IIL| 41-1 |g. ST] pouspavyy SIL| LT. 1\g- stipertddns sy O9T| S&-T 0-91] peuoeprey, «¢ GJT| Gg. Tig. ZT/pomddns sy|**** °° Le) Ze |(s) &. T |e. 91 . L¥Z \(S)F-T|0- OT x a4 (9 (79 pouspre yy “ 4 « | porsouuy Tee |(s) 9. T|4- ST a O8F| G6. L/S. 02| potopare yy OS? | S6-T\&- 02 perddns SV LLL \(8) &- & |0- 9T 0$9| PI-€|8- OT 2 66 16-0 |G. 9T go | 16-06-11 a 998 |(s) £-3|0- 9T c ecg |(s)9. ZO. 9T| pouspaeyT 06S Brg |L- Gt) pepwounty. SEG |Z. ST 998] 0-2 /6- $1] powopxoH ‘mu | “p | 7@ | woryrpuog Ty aan: | 0 (73 (73 ce ce 66 “et "T9999 29y00NT ee oe ae 91”990 sé (3 ** 79048 wagsbun 7, a3 ‘jo048 asaunhunyy "19098 ,, JOUSVT 5, ‘uowtoodg IR[MIAIN B PLY SIOY}O TTP °*92772 UT poANS St YONS Fo oer Teuorsuotmrp oy, foarenbs st Mommo0ds O]} JO WOT}OOS SSOTO OTT] FLT[Z SOTFIUGIS (S) x mi AN oD NAINA oes Do OF DAAGHN ) H190 OM ODD ¢ Lo aa en tA AN fe) Zi 214 Mr. J. R. Ashworth. Methods of making Magnets magnetised at the air temperature; No. 7 was made red hot and allowed to cool whilst in the magnetic field. No. 8, a specimen of cobalt steel, was kindly supplied by Mr. Hadfield, and is probably unique. All of these and the first example of nickel steel, No. 9 on the list, are Sheffield steels. Attention was then directed chiefly to three classes: Nickel steels, cast irons, and steel pianoforte wires. Nickel Steels —The first of these, No. 9, is a crucible steel from Sheffield, containing 3 per cent. of nickel and about 0°45 per cent. of carbon. The next two are from Scotland. They contain 2°4 per cent. of nickel and 0:19 per cent. of carbon. Nos. 12, 13, and 14 are also from Scotland, and contain 3 per cent. and 27 per cent. of nickel. They were kindly supplied by Mr. Riley, of the Glasgow Iron and Steel Company. The behaviour of the last three was remarkable, as when hardened they exhibited a small, negative coefficient. On heating and cooling they continuously lost magnetism for the first three alternations; at the fourth and fifth heating and cooling there was hardly any change of intensity; afterwards a small increase of intensity with rise of temperature and decrease with fall of tem- perature regularly took place. In the specimen containing 3 per cent. of nickel these operations caused a total loss of no less than 50 per cent. of the original magnetic intensity. This same piece was then annealed and magnetised ; the coefficient was now positive, the intensity rather higher, and the total loss 30 per cent. On re- hardening the events first described were reproduced, the negative coefficient and large total loss being almost exactly as before. It is very likely that by carefully adjusting the degree of hardness in this kind of steel a zero coefficient could be obtained. The 27 per cent. nickel alloys, after hardening in cold water, became almost non-magnetic, as discovered by Dr. John Hopkinson,* and it was only in this state that they were tested. No. 13 was magnetised at the air temperature; No. 14at —16°. All the other examples of nickel steels had positive coefficients. Cast Irons.—Specimens of grey cast iron, as used for general castings made at different times and of different blends of pig irons behaved very similarly. Magnetised as supplied they did not take a high intensity, lost permanently 30 to 40 per cent. of their magnetism, and had a large temperature coefficient. When hardened their magnetic properties were very different ; the intensity was then com- parable with that of tungsten steel, the total loss only about 15 per cent., and the temperature coefficient as low as, or lower than, the best examples of hardened steels. In three different kinds of care- fully hardened cast-iron magnets it was from 0:00016 to 0:00018 per degree centigrade. The average value for steel magnets of a similar * Hopkinson, ‘ Roy. Soc. Proe.,’ vol. 48, p. 61. independent of Changes of Temperature. 215 size tested at the Kew Observatory is given by Whipple as 0:00029.* The change of intensity with temperature is almost strictly linear in these cast-iron magnets, and they are very constant when subjected to blows and shocks. Pianoforte Wire-—Lengths of 12 cm. each were cut from a coil of wire, and tested after various treatments. Magnetised in the normal state this material unexpectedly gave a negative coefficient. When heated to bright redness and chilled rapidly or slowly the coefficient became positive. : As it was thus possible to change the sign of the coefficient, an attempt was made to find the particular temper which would give a zero coefficient. Lengths of the wire were heated severally in oil to 200° and 260°, and in air to a temperature producing a film of oxide, and rapidly chilled in water. The coefficient still remained negative, and of nearly the same magnitude. But when heated to dull redness and quenched, the coefficient was very nearly reduced to zero. Heated to higher temperatures and quenched, the coefficient became positive. Table IT. No. Condition. R=2i/d. Ti. ie B. «0-00. 16a | As supplied ...... 109 649 °O0 6446 0°008 | —0:'023 Tempered at 260°.. : 7o2°% 769 °2 0°029 | —0:018 Ditto dull red .... 2 8830 | 863°6 | 0-022 | —0-002 DPGLOs, GUGGO o¢. 0:00 sie a 892 °0 869 ‘0 0°026 +0003 Glass hard ....... ie 559 °5 537 1 0-040 +0008 oot CT 3 849-0 830 °1 0-023 +0006 163 | Assupplied.......] 100 679°0 | 633°6 | 0-067 | —0-055 Glass hard ....... es 593 ‘0 497 -O 0°163 —0:017 Length of each piece, 12 cm.; weight, about 09 gram; diameter, 16a = 0°11 em., 166 =012cm. These two specimens are made from different kinds of steel. It is a curious coincidence that the intensity of magnetisation attains a maximum for the condition producing minimum tempera- ture coefficient, and this maximum has the exceptionally high value of 892 C.G.S. units. The fact that the negative coefficient could not be reproduced if once the wire had been heated above a red-heat indicates that there is some structure physically imposed upon music wire, perhaps in the process of drawing, which partly or wholly contributes in pro- ducing the negative coefficient. Whereas the negative coefficient in the nickel steel is reproducible, and is doubtless a consequence of * Whipple, ‘ Roy. Soc. Proc.,’ vol. 26, p. 218. 216 Mr. J. R. Ashworth. Methods of making Magnets intense hardness. In contrast with this it may be mentioned that music wires are not at all hard, being easily touched with a file. In order to gain further insight into the cause of the negative coefficient in these wires, some experiments were made to test the effect of removing successively the outer layers of the wires by dissolving them in nitric acid. This revealed the important relation that the coefficient became more negative as the diameter became less, the length remaining the same, that is to say, as the dimension ratio increased. | To verify this carefully a series of stout music wires of different thicknesses, but in other respects as uniform as possible, were pro- cured from a manufacturer at Warrington, to whom I am also indebted for kindly supplying other samples of steel wire. The results of these experiments are most conveniently exhibited in tabular form, and are here annexed. Table IIT. axd. No. d. mM. R. Ii. IFS a. 0-00 0 -0000 33 | 0°216 | 3°535 55°3 | 5380°3 | 428°5 | 0°192 —0186 | 294 30 | 0:187 | 2°590 | 63°8 | 592°8 | 508°6 | 0°142 — 0184 344, 28 | 0°174 | 2°285 69°'0 | 682°4 | 551°5 | 0°128 — 0226 393 L426 | 0°153.)1.°760 4 78°: | -736°0 | 65278 | Oars — 0203 310 | 24 | 07134 | 1°365 89:0 | 742°0 | 686°3 | 0°075 — 0306 410 Length of each piece = 12 em. With the exception of No. 26 (and No. 26 was anomalous in some other respects) the coefficients become progressively more negative as the dimension ratio increases. The increasing product of the co- efficient into the diameter shows that the coefficient changes more rapidly than the dimension ratio. The table also shows the regular diminution of the permanent loss, 8, and increase of intensity as the dimension ratio increases, relations which hold in further experiments of the same kind to be described later on. Several of these wires after being thus tested were dissolved in nitric acid, and the temperature coefficient determined at successive stages of the process without any remagnetisation of the wire. The results of No. 33 alone are here given, as they sufficiently exemplify what generally takes place under these circumstances. The negative character of the coefficient progresively increases with increase of dimension ratio, and at a rather greater rate as in Table III. It is interesting to observe in these experiments the increase of independent of Changes of Temperature. | 217 Table IV. g No. d. m. R. f 33 0:°216 | 3°535 | 55°3| 5380°3 | 428 3) 192|—0°0136 | 294 | i{e stage| 0°195*| 2°875 | 61°3| 478°0 | 464°6 9 0: 0:°028/—0°0155 | 302 | 0:012!|—0°0196 | 319 | 0° 2ndstage| 0°163*| 1°995 | 73°6| 474°5 | 468° 005 | —0 °0292 ; a 8rd stage| 0°112*| 0°935 | 107°5| 485°5 | 482°9 mI Dissolved. intensity each time the wire is redissolved, remembering that after the initial magnetisation the wire was not subjected to any further magnetising process. Thus, for example, No. 33 has an intensity, after being heated and cooled, of 428; upon dissoiving off an outer layer the intensity rises to 478, which in its turn is reduced by heatings and coolings to 465; dissolving it a second time raises the intensity to 475, and so on. The recovery of magnetic intensity after dissolving in acid is most likely to be ascribed to diminution of the self-demagnetising force resulting from increase of dimension ratio. The intensities, however, after each dissolving, namely 478, 475, 486, are sufficiently constant to indicate that the intensity is nearly uniform throughout the wire, and this confirms an experi- ment of Bouty’s.+ The next two wires have been grouped in a separate table from the others, as they came from a different factory, being made in Sheffield. They are thicker than the former wires, and the thicker of the two, No. 34, has now a positive coefficient. By continually reducing the diameter of this wire, the coefficient ultimately changes sign and becomes negative. Table V. | No. | d. Mm. | iy I;. I. B. 0-00. 32 ‘| 0-227 | 8°875 | 52-8 | 490-1 | 340-6 | 0-305 |—0-0015 34 0°262 | 5:145 | 45-8 | 388-6 | 271-9 | 0-304 |+0°0220. Ist stage...| 0°223t| 3-740 | 53-7 | 307-2 | 299-2 | 0-026 |+0-0193 7 8 2 2nd stage ..| 0°204f/ 3°130 | 58° 297 °9 | 290°3 | 0:025 |+0°0184 3rd stage ..| 0°152t| 1°742 | 78° 3066 | 300°2 | 0-021 |+0-0082 | 4th stage ..| 0°075{t| 0°427 | 159° 292°5 | 287°1 | 0:019 |—0°0100 | Dissolved. Length, 12 cm. * Calculated from the weight. + ‘Ann. Scient. de Ec. Norm.,’ [2], 5, p. 131. ft Calculated from the weight. VOL. LXII. R 218 Mr. J. R. Ashworth. Methods of making Magnets And it may be calculated that if No. 34 had just been dissolved so far as to have a dimension ratio of about 110 to 115, it would have exhibited a zero coefficient. Since the former series of wires with dimension ratios of this magnitude would have had large negative coefficients, there must be some important physical or chemical differences between these and the former wires influencing the character of the coefficient. To complete the series of experiments on the influence of the di- mension ratio it was desirable to perform the converse operation and to prove that an originally negative coefficient would become positive by increase of thickness. Three pieces, (a), (b), (c), of No. 33 wire were cut from the same coil, each 12 cm. long, magnetised and then heated and cooled sepa- rately inthe same way. The coefficient was about —0:000119 for each. (w) and (b) were then bound together with fine copper wire, like poles being in contiguity; the coefficient as now determined was almost zero. ‘The piece (c) was then joined in the same manner to its two fellows and the coefficient again determined ; it was now + 0:000105. The experiment is conclusive, for it is allowable to regard bundles of wires as rods of equivalent cross section.* Wires drawn to different thicknesses are not structurally suffi- ciently identical to allow of strictly comparable magnetic results. It is therefore more satisfactory to vary the dimension ratio by altering the length and keeping the diameter constant. A series of tests were conducted in this way. Lengths of 3, 6, 9,12, 15, and 18 cm. of No. 30 wire were cut from the same coil, separately magnetised, and the coefficient of each very carefully determined. Table VI gives a complete view of the results. Table VI. : a. No. 21. R. ii. f B. 0-00. se a — EEE ——— enn ene oe | eee | eee 30 3 cm, 15 95 137 °4 18 °7 0:°427 + 0261 \ 6 ,, 31:90 313 °4 204-0 0°349 | +0151 q ae AT *85 483 °2 3783 0:217 — 0084 as 12) 5, 63 °80 602-0 513°8 0°147 — 0225 i 15 ,, 79°75 683 °1 595 °0 0:129 —0296 aS el 53 95 70 726°8 637 *4: 0°123 —0317 Diameter of each piece, 0°187 cm. * Von Waltenhofen, ‘ Wien. Ber.,’ vol. 48, part 2, p. 578, 1868. Ascoli and Lori, ‘ R. Accad. dei Lincei,’ Rome (5), 3, 2 Sem., p. 157, 1894. independent of Changes of Temperature. 219 The coefficient changes from positive to negative between the jengths 6 and 9cm. And hence if the change between these points is nearly linear, a length of about 8 cm. should have a zero coeffi- cient, and it might also be calculated that the permanent loss would be 0°262. A fresh length of exactly 8 cm. was cut from the same coil of wire and was found to have a coefficient of —0°000015, and a permanent loss of 0°281. A piece of this wire, a very little less than 8 cm, long, would without doubt, have a strictly zero coefficient. There are thus two practicable ways of obtaining zero tempera- ture coefficients, either (1) by altering the hardness, or (2) by altering the dimension ratio; and the latter may be effected by varying the diameter for a constant length, or the length for a con- stant diameter as may be the more convenient.. In addition, the material of which the magnet is made must have certain chemical and physical properties, not yet determined, of which, as far as some experiments I have made can decide, the physical rather than the chemical properties are the more important. Some of the results in Tables IV, V, and VI are here plotted as curves and exhibit interesting features. The curve of the relation of coefficient to dimension ratio (diameter constant) from the data of Table VI, Diagram I, curve (1), has a deuble inflexion between which it crosses the axis of abscissee and at either end apparently approaches to horizontal asymptotes. ‘This curve is probably typical of the behaviour of music wires. Curve (2) on this diagram traces the series of experiments on No. 33 wire. The two first points on the left correspond to the R 2 220 Mr. J. R. Ashworth. Methods of making Magnets coefficients for three and two pieces bound together, the third point: that for a single piece, and succeeding points the coefficients for the same piece at three stages of dissolution. The third curve is con- structed from the data in Table V, and represents the passage from a positive to a negative coefficient in No. 34 wire. 4. Ni “iS MN aN ) as Qi Si PAM ANENL, Diagram II exhibits the curve of permanent loss, 8, and dimension ratio, R, taken trom Table VI for No. 30 wire, diameter constant,. and. it will be seen it follows remarkably closely the path of the co- efficient curve. The coefficient, «, and the permanent logs, £, may then be connected by a linear equation | a=—at+bf. . S ‘The values of the constants for this material are a@ = —0'0005228-+0'0000073 and b = +0°001886 +0:000043. If curves for « and f be plotted with demagnetising factors, 7.c., the demagnetising force per unit intensity, corresponding to their dimension ratios as abscissee they resemble, strikingly, curves of magnetisation, having a point of inflection near the beginning and ultimately approaching horizontal asymptotes (Diagram III); by prolonging the curves in this diagram until they cut the axis of independent of Changes of Temperature. 221 ordinates it is easy to estimate what may be called the “ character- istic” temperature coefficient and permanent loss for this kind of wire. It may be inferred that in general the temperature effects upon magnets are principally influenced by the demagnetising factor over a considerable range of dimension ratios, and beyond that range by the nature of the material. In the fourth diagram the curves of initial and final intensities are plotted with dimension ratios as abscisse, and they resemble so closely the curve traced in the same way by Barns* for steel of “blue annealed” temper, that it is very probable this is the temper given to the music wires upon which these experiments have been made. The chief points elicited by this investigation may now be sum- marised :— | 1. The temperature coefficient is generally least in the hardest D1 agram-IT ‘71, i yas | fy, Lire Temperature Coefficrcnt ! Muaynetic Loss Permanese | Demagnet Ca Oi Pi ao er iplo TT 617 ia! "10080 60 50 90 70° * Barus and Stronhal, ‘ Bulletin U.S. Geol. Survey,’ No. 14, 1885. 222 Methods of making Magnets independent of Temperature. “4 Ni ~ IN N) | 2 £Fy4 ~~ PS IN}, a) —2- Ng} = % NIE --™ | ! | | | | | ye | A Tt Dimension Fiatio, 20 40 60 irons and steels, and is particularly small in hardened cast iron- Certain hardened nickel steels have very small negative coefficients. 2. The discovery of negative coefficients in music wires. 3. Change of the sign of the coefficient by alteration of (a) temper and (b) dimension ratio, and hence methods of obtaining zero co- efficients. 4. Some relations hetween the dimension ratio and self-demagnetis- ing factor, temperature coefficient, and permanent loss of magnetism after alternate heatings and coolings. An important consideration in any practical application to mag- netic instruments of magnets with zero coefficients is the constancy of the zero state. It is not yet possible to speak precisely on this point, but two wires which had been prepared by adjustment of temper to have zero coefficients in June, 1896, and since then had been lying on a shelf, and in the vicinity of other magnets, when tested nine months later, had not altered so much as to have a coefficient of practical The Electric Conductivity of Nitric Acid. 223 consequence. The intensity had diminished, however, by nearly 25 per cent. Similarly the magnet which had been given a negligible coefficient by cutting the length of the wire to 8 cm., as cited above (p. 219), after being boiled at intervals for four hours, was found five months later to have changed so little that its coefficient might still be con- sidered negligible. Further experiments, however, upon this question and some others arising out of this investigation are now in progress. “The Electric Conductivity of Nitric Acid.” By V. H. VELEY, M.A., F.R.S., and J. J. MANLEy, Daubeny Curator of the Magdalen College Laboratory, Oxford. Received Novem- ber 1,—Read December 9, 1897. (Abstract. ) In this paper an account is given of determinations of the electric © conductivity of nitric acid of percentage concentrations varying fom 1:3 to 99°97, purified, so far as possible, from reduction products of the acid, as also from sulphuric and the halogen acids, with which it is likely to be contaminated from its process of manu- facture. In the preliminary experiments it was observed that the results might be vitiated by (i) a trace of nitrous acid either directly added or produced by decomposition due to exposure to sunlight, and (ii) imperfect insulation of the electrolytic cell caused by metallic clamps, a point which seems to have been neglected by previous observers. The methods adopted for the purification of the water and nitric acid, as also for the detection and estimation of the impurities, are described in full. The greatest quantity of nitrous acid, sulphuric acid, and the halogen acids found in any sample used were 0°75, 4°3, and 3°8 parts per million respectively. The thermometers, resistance coils, and other instruments used were compared with certain standards and corrected accordingly ; the burettes and electrolytic cells were calibrated by one or more methods, and the mean of the values accepted. The method adopted for the determinations was in outline that originally described by Kohlrausch, but modified so as to overcome certain difficulties experienced. A particular form of bridge was constructed, in which the wire was an air line, and a special form of slider adopted to tap without sagging the wire, so arranged that it could be moved by the observer from the extremity of the bridge, and thus ail thermo-currents due to his proximity were avoided. 224 The Electric Conductivity of Nitric Acid. A rapidly revolving commutator was substituted for the usual induction coil, as the latter was found to be unsatisfactory owing to the susceptibility of nitric acid to polarisation. Various forms of electrolytic cells were used according ta the con- centration of the acid and the temperature of the observations; these were provided with movable electrodes, so as to throw into circuit different lengths of acid. A special form of apparatus was devised to prepare nitric acid of 99°88 per cent., and another form to obtain acid of 99°97 per cent. from the latter. As a considerable quantity of this practically anhydrous acid was obtained, its chemical and certain physical pro- perties were examined. It has no action on (i) copper, (ii) silver, (iii) cadmium, and (iv) mercury, all of high degree of purity, and (v) commercial magnesium, at ordinary temperatures; purified iron and commercial granulated tin were unaffected by the acid, even when boiling. Purified zine was slightly acted upon, but sodium immediately caught fire. The acid has no action whatever on calcium carbonate at ordinary temperatures or the boiling point. Flowers of sulphur and iron pyrites dissolve quickly and completely in the gently warmed acid. The following results were obtained for the density of the 99°97 per cent. acid, corrected for weighings in VACUO :— Density 4/4 = 154212; 14°2/4 = 152234; 24-2/4 = 150394, the mean values of two concordant observations. As a further check upon the measurements obtained by the Kohlrausch method, certain other measurements were made by Carey Foster’s method for the comparison of resistances, and the results obtained were found to be concordant within the limits of experimental error. Inaseries of tables the values are given for thirty-two samples of acid of the specific resistance in true ohms at temperatures of O°, 15°, and 30°, the temperature coefficients «10+ and $10® deduced from the equation R; = R,(1+at— fr’), as also for Ky x 10°, Ky; x 10%, and K,)x10* (the conductivity of mercury at 0 being taken as unity, and its specific resistance as 94°07 microhms ver 1 c.c.). It is shown that the specific resistance decreases for percentage concentrations from 1°30 to 30, at first more, then less rapidly (thus confirming the previous observations of Kohlrausch) ; from this point the resistance increases slowly up to 76 per cent., thence more rapidly until a maximum is reached at 96°12 per cent., when a sudden reversal takes place. Further, whereas nitric acid behaves as other electrolytes in pos- sessing a positive temperature coefficient of conductivity for percent- age concentrations from 1'3 to 96:12, yet from this point up to Refractivities of Air, Oxygen, Nitrogen, Argon, Gc. 229 99:97 per cent. it behaves as a metallic conductor in possessing a negative temperature coefficient. Similar phenomena have been observed by Arrhenius in the cases of moderately dilute solutions of hypophosphorous and phosphoric acids, and explained by him by means of the ionic dissociation hypothesis. It is pointed out that nitric acid of 96—99:97 per cent. would ex hypothesi contain few, if any, free ions, and therefore the theory would lead to a totally opposite conclusion. The results of the experiments are also discussed in relation to the hydrate theory of solution, and the illustrative curves in which the percentages of acid are taken as abscisse and the resistances or con- ductivities in merenry units show points of discontinuity markedly at percentages corresponding approximately to the composition required for the hydrates HNO;,2H,0, HNO;,H,O, 2HNO.,H,O (= H,N,0;), and less markedly for the hydrate HN O;,10H,0. Further, if the values of « x10‘ and $x 10° are referred to molecular proportions of water, the minima values of the former and the maxima of the latter occur in the cases of 3:07, 1:84, 0°99, and 0°55 molecular proportions or very approximately HNO,,3H,0, HNO;,2H,0, HNO,,H.O, and 2HNO,,H,O. Further evidence is thus added by an independent method to that already accumulated as to the existence of definite combination of nitric acid with water. Finally, it is pointed out that if a curve is plotted out in which the molecular proportions of water are taken as abscisse and the values for «10* as ordinates, there are ascending and descending branches, meeting at the points corresponding to the formation of the respective hydrates ; the phenomena are compared with those observed by Bakhuis- Roozeboom for the solubility curves of hydrates of ferric chloride and by Le Chatelier, as also by Heycock and Neville for the freezing point of alloys. “On the Refractivities of Air, Oxygen, Nitrogen, Argon, Hydrogen, and Helium.” By Professor WILLIAM Ramsay, Ph.D., LL.D., Se.D., F.R.S., and Morris W. TRAVERS, B.Sc. Received November 18,--Read December 9, 1897. In the course of a research on the nature of helium many measure- ments of its refractivity referred to that of air as unity were made by means of an apparatus similar to that described by Lord Rayleigh.* Inasmuch as the refractivity of helium is very small it was not found convenient to measure its value directly against air; hence it was compared with hydrogen, and hydrogen was compared with air. * Proceedings,’ vol. 59, p. 203. 226 Prof. W. Ramsay and Mr. M. W. Travers. And as acheck on these measurements, the hydrogen was compared with oxygen and subsequently with nitrogen free from argon. It was noticed, after some of these experiments had been made, that the refractivity of air could not be accurately calculated from the given data for oxygen, nitrogen, and argon; and it appeared therefore worth while to examine more minutely the refractivity of these gases for white light, and to see whether any error could be detected in previous measurements. Moreover, as physicists perhaps do not always devote sufficient care to the chemical purity of their mate- rials, an additional reason was furnished for the inquiry. Apparatus.—It will be seen, on consulting Lord Rayleigh’s paper, that the refractivity is measured in the following manner :—Light from a parafiin lamp passes through a fine slit, cut with a razor in tin-foil pasted on glass. The beam is made parallel by passage through an achromatic plano-convex lens of about 1 foot focal length. It then divides; the upper portion passes through air, and, after extraneous light is cut off by passage through two wide slits, it is brought to a focus by a lens similar to the first, and the bands produced are viewed by acylindrical lens of very short focus. The lower portion of the beam traverses two tubes, 9 inches long and one- quarter of an inch in diameter, piaced close together, and closed at each end with plates of optically worked glass. Hach of these tubes contains one of the gases to be examined; and each is connected with a manometer and a movable reservoir; so that, on raising or lower- ing the reservoir, the pressure of the gases can be so adjusted that the interference-bands formed in the lower half of the field can be accurately brought into line with the stationary bands in the upper half. Readings of pressure are taken on both manometers at pres- sures not differing greatly from that of the atmosphere; then, on lowering the reservoirs, readings on both manometers are taken at lower pressures, the bands being again made to coincide in position with the upper fiducial bands. The ratio of the refractivities is inversely as the differences of pressure in the two gases. The in- fluence of temperature does not appear, for the tubes of the mano- meter lie side by side, and may be regarded as equally affected by variations of temperature. The accuracy of this method varies with the value of the refrac- tivity of the gas. For, if the gas has a low refractivity, then a great difference of pressure produces the passage of fewer bands across the field than if it has a high one; and, as the accuracy of reading may safely be taken as the twenty-fifth of a band, and as between thirty and forty bands passed the field with such gases as oxygen, nitrogen, and argon, the error may be taken in such cases as from 1 in 750 to 1 in 1000. The tubes containing the gases to be examined were connected Refractivities of Air, Oxygen, Nitrogen, Argon, §c. 227 with a Topler’s pump; and before admission of gas each tube was pumped empty, so that in an attached Plucker’s tube there was brilliant phosphorescence. The tubes were then washed out with the gases to be admitted, the apparatus again evacuated, and the final quantity of gas allowed to enter by a contrivance a description of which is to be found in the ‘ Trans. Chem. Soc.,’ vol. 67, p. 686. Purity of the Gases. Hydrogen—The hydrogen was made by ‘warming a tube containing palladium-hydrogen which had been pre- pared by admitting hydrogen made from pure zinc and sulphuric acid into contact with spongy palladium. The tube was pumped empty in the cold, and then gently warmed; it was again allowed to ‘cool and again pumped empty. The hydrogen was then coilected, passing slowly through a tube, filled with phosphoric anhydride, into the experimental tube. ' Oxygen—The oxygen was prepared by heating a small tube con- taining potassium permanganate; a large quantity of gas was allowed to escape, and a portion was collected finally which served for the experiments. Nitrogen.—The nitrogen was prepared from a mixture of am- monium chloride and sodium nitrite, to which a little copper sulphate had been added. The apparatus was exhausted before admission of either of the solutions, and before allowing the solu- tions to enter they were boiled, and the flasks corked while boiling. The gas was passed over red-hot copper; the ammonia liberated by the alkalinity of the nitrite thus reacted with any oxides of nitrogen possibly present to form water. The gas was collected, after rejec- tion of a considerable portion, in a tube containing oil of vitriol; it was then transferred to a fresh tube, treated with a very strong solu- tion of caustic potash, and finally admitted to the apparatus. Air.—The air was left standing for some hours in a tube con- taining sticks of caustic potash, and was then admitted to the appa- ratus through a tube of phosphoric anhydride. Experimental Data.—Each gas was compared with air and with the other two. Ai§r is in each case taken as unity. Hydrogen.—Hydrogen/air ...... 04730 0:473'7 Mean.. 0°4733 Hydrogen/oxygen .... 0°5125 peor: 0°5125 9 eo 0) 5125 Hydrogen/nitrogen.. 0°4654 ABD A 0-4054 Bae 0°465 Oxygen.—Oxygen/air......... 0°9237 — -0'9262 ee. W'OedS 0-9230 228 Prof. W. Ramsay and Mr. M. W. Travers. Oxygen.—Oxygen/hydrogen... 1°9512 19512 Oxygen/nitrogen.... 0°9090 0°9122 tae ae eOS 0-9103 Nitrogen.—Nitrogen/air........ 10153 10171 sreatiun SIO LGG 10174 Nitrogen/hydrogen.. 2°1487 2°1487 ) Nitrogen/oxygen.... 1:1001 10962 10986 Mean.. 1°9512 2°1487 at Geen DOSS To these numbers those for argon must be added. The gas was pre- pared in the usual manner from air; and before admitting it into the experimental tube it was sparked with oxygen in presence of caustic soda fortwo days. The cxygen was removed with phosphorus, and the argon, on its way into the experimental tube, passed over phosphorus pentoxide; a Pliicker’s tube was sealed to the tube through which it entered, so that its spectrum might be observed. It contained no visible trace of either hydrogen or nitrogen. Argon.—Argon/air ......c0ese0. 0°9596 0°9596 Mean.. 0°9596 ATF 00 | Oy Sein a cys nets aids 10350 1:0348 silt tibet USD Argon/nibtOven ne sa. 0:9412 0-941 9 = 0°9416 Placing air in each case equal to unity, and calculating the refrac- tivities of the other gases, we obtain the following table :— Refractivities of Gases, Air equal to Unity. © Through eee Directly lame —~ compared. Oxygen. Nitrogen. Hydrogen. Argon. Hydrogen .... 0°4733 O-4737 O-4727 a — OXY CT 6 diss 0°9243 —— 0:924:7 0:9237 0°9261 Nitrogen .... 1°0163 1:0155 = 10170 10191 ATO a. nia «= 0°9596 0:9577 0°9572 — — Carbon dioxide — 1°5316 ons 22 cae Refractivities of Air, Oxygen, Nitrogen, Argon oe. 229 It will be seen on inspecting the above table that the numbers obtained indirectly are in close agreement with those obtained by direct comparison with air. Taking the value found directly by Mascart for D*, viz. (n—1l)p = 0'0002923, the value found by him for nitrogen (atmospheric) was 0°0002972, giving for nitrogen, on the basis air equal to unity, the number 1:0178. Mascart did not determine the value for oxygen, but calculated it from the above data and the known composition of air. Nor did Lorenz determine the value for nitrogen; but taking his own value for oxygen, viz. (n--1)y = 0°'000272, and for air (n—1)p = 0°000291, he deduced it, as Mascart had done for oxygen. So that we have no determination of the three constants, or their com- parison, by any one observer since Dulong in 1826. It has been tacitly assumed that the refractive index for a mixture of gases is that of those of their consituents, taken in the proportion in which they occur. We have in our hands a means of verifying this assumption, which is well known not to hold for compound gases, nor for mixtures of liquids, even though change of density be taken into consideration. Dulongy gives very careful accounts of the methods he used in preparing the samples of gas that he employed. Oxygen, to which he ascribed the refractivity 0°924, was obtained by heating potassium chlorate. His result is identical with ours. Nitrogen was prepared from air by absorbing the oxygen with phosphorus, first at a high temperature and then in the cold. It was then washed with a solu- tion of chlorine, and afterwards with potash. It is difficult to see what object was to be gained by washing with chlorine water, unless it was the removal of hydrogen. The number he obtained was 1:02, somewhat higher than that which we have found. Dulong also determined the refractivity.of air, and allowing for that of the small percentage of carbon dioxide, it is precisely the mean of that of its constituents, taken in the proportion in which they are present. Returning to the results of Mascart and Lorenz, we have for the D lines :— Air. Nitrogen. Oxygen. Miiseart 2.20.30 1 10178 —— oreng. 2 isis s 1 — 0:9347 From these data of Mascart and Lorenz it is possible to calculate the refractivity of air :— | (1:0178 x 79:1) + (0'9847 x 20°9) = 100°15. There is reason to doubt the purity of Lorenz’s oxygen. He heated * The dispersions for these gases are so small as not to affect the ratios of these numbers (‘ Compt. Rend.,’ 1874, vol. 78, p. 621). + ‘ Ann. Chim, Phys.,’ vol. 31, p. 176, 1826. 230 Prof. W. Ramsay and Mr. M. W. Travers. mercuric oxide, of which he does not give the method of preparation ; it may have contained oxides of nitrogen; and for some reason, not explained, he passed the gas through a vacuous porcelain tube, pre- sumably red-hot, which, as recent experiments of Messrs. Bone and Jerdan have shown,* is not impervious to furnace gases. Dulong, on the other hand, who, as already remarked, prepared his oxygen from chlorate, obtained the number 0:924 for white light, coincident with our determinations. The refractive index of air, calculated from our determinations, VIZ. Wxyoen. 7S) 2 a. wee 0:9243 Natroren "s). 7o. see 1:0163 scorn S11, +, Ye aan 09596 and the densities of the constituent gases,} gives the following numbers :-— (1:0163 x 78°15) + (0:9243 x 20°91) + (0:9596 x 0:94) = 99-653. Observers sometimes find the percentage of oxygen in air to be about 20°98, or even 21:0. . This would hardly affect the result ; with 20°96 per cent. of oxygen the calculated refractivity is 99°647, fae of 99-653. There can be no doubt as to the refractivity of oxygen from our ratios, as well as from Dulong’s determinations. The question is as regards nitrogen. It would require the refractivity of nitrogen to be 1:208, a number greatly above any of our values, in order that the sum of the refractivities of oxygen, nitrogen, and argon should equal 100. The presence of argon would also make an almost inappreciable difference. Taking Tileercara determination of the refractivity of atmospheric nitrogen to be correct, that of pure nitrogen would be 1:0181, instead of 1:0178. And an error in the refractivity of argon would also not affect the result, inasmuch as the total amount of argon is so small. We are thus driven to conclude that the refractivity of the mixture, air, is somewhat less than that of the sum of the refrac- tivities of its constituents, taken in the proportion in which they occur. It appeared advisable to try other mixtures; and a mixture of hydrogen and helium was first selected, because these are both very ‘“‘nerfect”’ gases, inasmuch as their critical points lie very low. It was to be expected that if a difference between calculated and found values should exist, it should be of the inverse character to that of a * ‘Chem. Soc. Trans.,’ 1897, p. 42. + Argon, ‘ Phil. Trans.,’ A, 1895, p. 202, foot-note. Refractivities of Air, Oxygen, Nitrogen, Argon, §c. 231 mixture of oxygen and nitrogen, for they are two somewhat “ im- perfect’ gases. The result has borne out this idea. A mixture was made of 20°60 c.c. of hydrogen and of 20°12 c.c. of helium free from argon, and of the density 1:960; and with the refractivity of the mixture those of hydrogen and helium were com- pared. Taking the refractivity of the mixture as unity, the follow- ing ratios were found :—- Hydrogen/mixture ..... aby sSNA 1-5957 ean 1°5967 ’ Helium/mixture......... 0°4513 0-447 brant ‘@ienaey The calculated values are— (04495 x 20712) ... (1:5967 x 20°60) 80°87 A-72 = r02-59., Here the calculated value of the refractivity of the mixture is 3 per cent. higher than the found value, while with air the calcu- lated value is 0°35 per cent. too low. A third experiment was made, in which the “artificial air” was a mixture of 19°13 c.c. of carbon dioxide with 19°29 c.c. of oxygen, both gases supposed to be at 0° and 760 mm. Again, taking the refractivity of the mixture as unity we found the following ratios :— Carbon dioxide/mixture....... 1°2450 rayon) mixture... 0... ses 5 o Ofaee The paleolated values are :— (1:2450 x 19:13) (0°7525 x 19°29) 37-78 38°42 ee Here, as with air, the total refractivity found is less than that calculated. It is true the difference is not great, but we are per- suaded that it is real, for it considerably exceeds the error of our . several determinations. The case is not bettered if Lorentz and Lorenz’s formula be substi- tuted for Gladstone and Dale’s. Using their formula, 1?—1/n?+2, the calculated result is 99°72 per cent. of that found for air. The coefficient of compressibility of hydrogen is too small, while that of other gases, such as oxygen and nitrogen, is too great. The 232 Mr. E. J. Bles. On the Openings in the effect of mixing equal volumes of hydrogen and helium, each of which has too large a coefficient of elasticity, is to cause each to occupy twice the volume that they previously occupied, and to halve approximately the pressure for each. The pressure is therefore lower than it would be for an absolutely ideal gas, for each gas, hydrogen and helium. The sum of these pressures will accordingly be too low, or transposing, the sum of the volumes will be too great. The opposite argument holds for air. Now, in considering volumes we deal not merely with the co-volume, i.e., the space occupied by the molecules, but also with the inter- stitial space-inhabited by the molecules. But the refractive power, if Clausius’s deduction from the formula of Lorenz and Lorentz is correct, is a function of the dielectric constant, and hence of the co-volumes of the gases. And here the discrepancy is more easily detected than by any determination of density. Jt must therefore be concluded that gases are not, as postulated by Dalton, indifferent to one another’s presence, but that they modify one another’s properties in the same manner as do liquids, though to a different extent. This mutual action at high pressures and small volumes modifies even the volume relations, as recently shown by Dr. Kuenen. And it must persist at low pressures and large volumes, though it may not always be possible to make measurements of pressure and volume accurate enough to lead to its detection. The refractivity, however, seems to be a means delicate enough to be used for this purpose. “On the Openings in the Wall of the Body-cavity of Verte- brates.” By Epwarp J. Burs, B.Sc. (Lond.), King’s College, Cambridge. Communicated by Dr. HANs Gapow, F.R.S. Received June 16,—Read June 17, 1897. In the review of the vertebrates held in the following pages, I have put together as many facts as I could ascertain on the distribution of abdominal pores in the various groups, and side by side with this evidence I have arranged the available facts recorded by others, and observed by myself, on the distribution of nephrostomes and other openings on the wall of the abdominal cavity. By so doing, the physiological meaning of the abdominal pores has, I believe, been elucidated through the evidence of a correlation, speak- ing generally, of an alternative character, between these two sets of organs. It will further appear that in most of the higher vertebrates —where abdominal pores do not occur and nephrostomes disappear early in development or lose their original connection with the renal ducts—the body-cavity has taken upon itself a different functional character. Instead of acting as auxiliary to the excretory organs, it takes part in the internal work of the circulatory lymphatic system. Wall of the Body-cavity of Vertebrates. 233 The greater part of the information on the openings from the perivisceral cavity to the exterior in the Elasmobranchii is contained in two papers; one by Semper,* on the urogenital system of Plagio- stomes, the other, by Bridge, on “ Pori Abdominales of. Vertebrata.” Semper describes the persistence in certain Hlasmobranchs of a number of open segmental funnels on the peritoneal epithelium leading into the Malpighian bodies of the mesonephros. Such funnels occur in all Elasmobranch embryos, but usually close during develop- ment. Semper gives lists of species with and without nephrostomes when adult, and shows that their presence cannot be correlated with the presence or absence of other organs, among which he did not, however, refer to the abdominal pores. Bridge was the first to examine a number of Hlasmobranchs expressly to determine the distribution of abdominal pores amongst the species of these fishes. He states that it was “‘ not clear that the presence or absence of the pores can be correlated with structural variations in other organs.” I was led to compare Semper’s and Bridge’s accounts of the distribu- tion of nephrostomes and abdominal pores, and it at once became evident that their presence in Hlasmobranchs was, to a certain extent, reciprocal. A few discrepancies which appeared have been investi- gated, and my results, although they agree in the main with Bridge’s, differ from his in one or two important cases. A detailed discussion of these cases will appear elsewhere. The species which have come under my own observation are :— Carcharias acutus, Riippel, C. glaucus, L., Galeus canis, Bonap., Zyyena malleus, Risso, Mustelus vulgaris, M. and H., M. levis, Risso, Hexanchus griseus, Gm., Heptanchus cinereus, Gm., Scyllium canicula, L., S. stellare, L., Pristiurus melanostomus, Bonap., Cestracion philippt, Lacép., Spinax niger, Bonap., Scymnus lichia, Cuv., Centrophorus granulosus, Bl. Schn., Rhina squatina, L., Pristiophorus cirratus, Lath., Pristis zysron, Blkr., Rhinchobatus djeddensis, Forsk., Rhino- batus granulatus, Cuv., Torpedo narce, Risso, Narcine brasiliensis, Olf., Raja clavata, L., R. maculata, L., Myliobatis maculata, Gray, and Myl. sp. Table I contains all the species for which there are data respecting both nephrostomes and abdominal pores. It includes all the species investigated by Semper, excepting Lamna glauca, M. and H., and Temera hardwickii, Gray. In these two species the nephrostomes close, and it may be expected that they will eventually be found to possess abdominal pores. Table II is so arranged that the species with fapledetomes are _ * C, Semper, “Das Urogenitalsystem der Plagiostomen und seine Bedeutung fiir das der iibrigen Wirbelthiere,” ‘ Arb: Zool.-zoot. Inst. Wiirzburg,’ vol. 2 (1875), pp. 195—509. + ‘Journ. Anat. and Phys.,’ vol. 14 (1879), pp. 81—100. VOL. LXII. S 234 My. E. J. Bles. On the Openings in the placed together under A, and the species without them together under B. The species under B all possess abdominal pores. Under A, on the other hand, there is no such uniformity-as regards the pores. We have here a fairly complete series of species; beginning with forms without abdominal pores like Cestracion, passing to forms like Scy. stellare, which. acquire pores late in life and may occasion- ally fail to do. so; we then come to the Scy. canicula group, where pores are found at the stage of sexual maturity, but where they may be acquired still later or sometimes not at all; and, lastly, there is a fourth group, that. of Acanthias vulgaris, where the pores appear at an early age, towards the end of embryonic life, and seem to be invariably present. These four groups have one character in common: the nephrostomes remain open in the adult. In this they differ from the species under B, which close the nephrostomes early in development, and then, like Group 4 of Series A, and like- wise at an early age (Carcharias during fcetal life), open abdominal pores. To some of the species in Table Ino place in Table II can be assigned until more specimens of different ages have been examined. It is sufficiently obvious from the list of species in Series B of Table II that abdominal:pores are distributed without reference to oviparity or viviparity. Table I. ~ Hlasmobranchii. pegmee Abdominal pores tubes. : | | ee ill |Open.| Closed. | © Present. | Absent. Sub-Order SELACHOIDEI. | | Fam. 1. Carchariide. : | a. Carchariina. | | | Carcharius glaucus, L..... | ae + | Galeus canis, Bonap. .... | = + ! b. Zygeenina. | : Zygena pe Risso, ¢ | juv.. erveecosese ee | = = G: Mustelina. Mustelus mulgar ee, M and ts He - = + M. Te Ricco, Q peice as —_ + Triacis semifasciata, Girard = + Fam. 2. Lamnide. Lamna cornubica, Gm. ....... ~ + (Fam. 3. Rhinodontide.) Fam. 4, Notidanide. Hexanchus griseus, Gm., ? juv. + + | Heptanchus cinereus, Gm.,2 ..; + (or +) | _ Wall of the Body-cavity of Vertebrates. Table I—continued. Sub-Order SELACHOIDEI—cont. Fam. 5. Scylliide. Scyllium canicula, L., 6, 7 5) ae ee, 6 (bridge) ..j)....... Ditto, é (Marshall and ays Ditto, « 12 cases (E. J. B.) .. Ditto, 6 and e) . Ditto, 2 (Marshall and Hur ws | Ditto, 2 juv., (ditto) . e Seyllium stellare, L. bs wifi Pristiurus melanostomus , Bonap. Fam.6. Cestraciontide. Cestracion philippi, Lacép., 2 . Fam. 7. Spinacide. Centrina salviani, Risso, 2 .. crea granulosus, BL Schn. : Ditto, g jur. ia ae sina mo Seymnus lichia, Cuv. ......... Acanthias vulgaris, Risso..... Ditto, 12” fetus (Bridge) . Spinax niger, aay Eo i Ditto, 2.. ie» ibiee 6a. Fam. 8. Rhinide. Rhina squatina, L. ........., (Fam. 9. Pristiophoride. a Sub-Order BArorper. (Fam. 1. Pristide.) Fam. 2. Rhinobatide. Rhinchobatus djeddensis, er ce. ss. caia esc © Rhinobatus granulatus, Cuv. .. Fam. 3. Torpedinide. Torpedo narce, Risso ....... » marmorata, Risso .... Hypnos subnigrum, Dum., g.. Narcine brasiliensis, Olf. ..... Fam. 4. Rajide. eS es » maculata, Morntag., eases (E. J. B.) » punctata, Risso... » miraletus, L. . » batis, L. 2 “s marginata, Lacép. we », blanda, Holt and Calder- SL ee Fam. 5. Trygonide. Trygon brucco, Bonap......... ” pastinaca, VE ee Fam. 6. Myliobatide. Myliobatis maculata, Gray, 2 = Se a | | (! Segmental ° Open. | Closed.| Present. +t+t+++¢t4+4++4 + ? tet eeee + + tubes. | | 239 Abdominal pores. | Absent. + (5 cases) | — (2 cases) + + (6 cases) | — (6 cases) ~ (or +) (or +) + Tee ° kB | Se ++ Ft t+ett + tee +4 + + 236 Mr. KE. J. Bles. On the Openings in the Table I. A. Elasmobranchs with nephrostomes when adult. (1) Abdominal pores absent. Cestraciontide. Cestracion philippi, Lacép. Rhinide. Rhina squatina, L. (2) Abdominal pores absent until full-grown. (Whether they are then con- stantly present is not known.) Seylliide. Sceyllium stellare, L. Pristiurus melanostomus, Bonap. (3) Abdominal pores appear late (when sexually mature) amd may be absent. Seylliide. Sceyllium canicula, L. Spinacide. Spinawx niger, Bonap. (4) Abdominal pores appear early and are constantly present. Spinacide. . to 20°C. . Solution of Potassic Hydrate in water, 5 per cent. solution. . Solution of Rubidic Hydrate in water, 5 per cent. solution. . Amyl alcohol. . Hthylic ether. . Ethylic ether, pure and dry. . Ethylic alcohol. The change in capacity of the sliding condenser when ethylic alcohol replaced the air in the experimental condenser was 16:7. Hence 8, = 16°7 and Dy = 25°8, also s = 1°36. OH Om OO DW Do—1l 248 heref eos eee = = 156138. Therefore So = TBA The following table shows the observed values of s in the several cases when the above liquids were placed in the experimental condenser, and - the corresponding calculated value of the dielectric constant D, where D = 1:613x (S—s) +1. of Organic Bodies, Se., at very Low Temperatures. 253 Table I—Determinations of the Dielectric Constants of certain Liquids at Ordinary Temperatures (15° C.) by Nernst’s Method. Frequency = 320. Dielectric constant Substance. — S—s. 1°613 (S—s). == 1) Kthylic alcohol (taken 16 7 15°34 24:8 25°8 as the standard of (assumed value) comparison) yi aicoho! ....... 10°49) 913 14:7 15°7 (calculated) Hthylic ether..... veg vader 2°62 4°23 5°23 | (calculated) Pure dry ethylic ether 3°70 2°34 3°78 4°78 (calculated) Hence by Nernst’s method, assuming the dielectric constant of ethylic alcohol to be 25:8, we find that of amyl alcohol to be 15°7 and pure ethylic ether to be 4°78. Nernst himself found amyl alcohol to be 16() and ethylic ether to be 4:25 at about this temperature. Hence our values are in fair agreement with his. In the next place we cooled the experimental condenser down to the temperature —185° C. in liquid air, after filling it with one of the above six dielectric liquids, and we repeated all the above- described operations again. The results are collected in Table II. Table 11.—Determinations of the Dielectric Constants of certain Frozen Liquids at the Temperature of Liquid Air by Nernst’s Method. Frequency = 320. Calculated J] ‘613 x dielectric Substance. S. S—s. (S—s). constant = D. Hthylic alcohol .... 2°68 1°32 2°13 313 mum@yl alcohol ...... 2°34: 0:98 1:58 2°58 Hthylic ether ...... 2°16 0°80 1:29 2°29 © per cent. solution of Rubidic Hydrate 2°94 1°58 2°55 oOo 5 per cent. solution of Potassic Hydrate 5°15 3°79 6:12 712 The above values for the organic bodies are in close agreement with the results we obtained for the same substances by the galvano- meter and switch method formerly used by us, as may be seen by a reference to Table III. 254 Profs. J. Dewar and J. A. Fleming. Dielectric Constants Table ITI.—Comparison of the Determinations of certain Dieleotric Constants made by different USA: at the Temperature of Liquid Air. By galvanometer By Nernst’s bridge and switch method. method with telephone. Frequency = 120. Frequency = 320. Substance. Dielectric constant. Dielectric constant. Bithylic alcohol var. e 311 3°13 amy) alcohol, <1. 0,598 e.= 2°14 2°58 Hthyliefethiony... thee sc). 2°31 2°29 5 per cent. solution (aqueous) of Potassic Hydrate...... 123-0 712 5 per cent. solution (aqueous) of Rubidic Hydrate ...... 816 Bi )y) The results collected in the above Table III, show that the two methods give practically identical values for the two alcohols and the ether, but very different value for the two frozen dilute hydrates An examination was then made of several other substances, and for this purpose another condenser was constructed, which consisted of a platinum crucible about 4 cm. in diameter and 5 em. high. This crucible was fitted with an ebonite lid, through which passed a glass test-tube, in the interior of which was placed our platinum thermometer. Round the outside of the test-tube, platinum wire was closely wound, so as to form the opposed surface of a condenser in relation to the platinum crucible as the other surface. This platinum condenser could then be filled with any electrolyte or organic liquid and frozen in liquid air. Owing to the very small actual capacity of this last experimental condenser, and especially that of the variable part of it in comparison with the capacity of the leads and connections, no very great accuracy of measurement was looked for or attained. The results, however, were sufficient to check the general jaccuracy of the experiment with similar substances by the galvanometer method. This platmum condenser was calibrated and used with the Nernst bridge, exactly as in the previous experiment. With the experimental condenser empty the change in capacity of the variable sliding condenser in the bridge arm was 1°50 on chang- ing over the position of the experimental condenser. Hence s= 1°50. When filled with ethylic alcohol (Dp = 25:8) the change of capacity of the sliding condenser was 6:20. We have, therefore, Sp = 6°20, D,, == Ao; ous =o): Therefore Doz an Ae So—s 4°70 The experimental condenser was then filled with some liquid, either at ordinary temperature or frozen at a low temperature, and Sood of Organic Bodies, §e., at very Low ‘Temperatures. 255 the bridge measurement made, and the reading S or the change of capacity of the sliding condenser observed, as before, when the experimental condenser had its position changed. The following Table IV gives the summary of the results obtained with several substances at various temperatures. Table IV.—Measurement of the Dielectric Constants of various Substances at different Temperatures by Nernst’s Method. Fre- quency = 320. S= 1°50, ae = 527. So—s Calculated Temperature dielectric in platinum Substance. S. S—s. 527(S—s). constant =D. degrees. Ethylic ether... 21 0°6 3°16 4°16. + 15° Glycerine ..... 12:15 10°65 56:2 57°2 +30 Solution of am- ¢ 4°5 3°0 15°8 16°8 —123 (2) monia (sp. en 5°6 ArT 216 22°6 —137 Stee... Liss 12°70 63:0 64:0 —119 ee 0-2 0°5 2°6 3°6 —49 Distilled water jn aie Stalk my ae oe (43 12-8 | 67% 68-0 +1 ee P| 205 245 (12-9 ee wet he ((12'8).113 . 59:7 60°7 —4] im water.... The value found for the dielectric constant of water at +1° is rather low, but, as above mentioned, the smallness of the capacity of the experimental condenser prevented the results from being more than good indications of the order of the dielectric constant. We have, in addition, repeated and extended experiments made with the cone condenser on various electrolytes and dielectrics. In the first place, we have carefully examined the effect of change of temperature on the dimensions of the cone condenser per se to ascertain if the dimensional change produced by cooling it in liquid air could sensibly affect the value of the dielectric constant of an electrolyte forming the dielectric between the cones, apart from the change which temperature produces in the dielectric quality of the dielectric itself. The gilt cone condenser was accordingly connected with the tuning-fork interrupter as formerly described,* and the galvano- meter scale deflection when the condenser was charged with 97:2 volts was found to be 3°85 cm. to the left and 3°88 cm. to the right, hence the mean galvanometer deflection was 3°87 cm. of the scale. This corrected to 100 volts becomes 3:98, and deducting 0:4 cm. for the capacity of the leads, gives 3°58 as the number representing the * See Fleming and Dewar, ‘ Roy. Soc. Proc.,’ vol. 61, 1897, p. 299. eT ore 206 ; Prots. J. Dewar and J. A. Fleming. Dzéelectrie Constants electrical capacity of the cone condenser as then arranged. A standard condenser belonging to the Davy-Faraday laboratory, and having a capacity of one-thousandth of a microiarad, was then sub- stituted for the cone condenser, and gave right and left deflections ot 1971 and 19°6 cm. respectively when charged with 97 volts. This corrected for capacity of leads (= 0'4 cm.) and reduced to 100 volts becomes 19°53. Hence, since the electrical capacities are proportional to the galvanometer deflection, the electrical capacity of our cone 3°58 condenser is 19-33 * 0 001 of a microfarad, or 0'000183 of a micro- farad. The capacity of the gilt cone condenser was then again measured with air at 20° C. as dielectric, and the galvanometer deflection observed. The outer cone was then cooled to —185° C. by quickly applying to it a large quantity of liquid air whilst the inner cone remained at about 20° C., and the galvanometer deflection again observed. This deflection was taken again when the inner cone had fallen in temperature to —75°, and finally when both inner and outer cones were at —185° C. The following numbers giving the galvano- meter deflections are then proportional to the electrical capacity of the condenser between the cones. Table V.—Examination of the Effect of Cooling on the Electrical Capacity of the Cone Condenser. Gaivanometer scale deflection in cm. or electrical capacity ot the cone condenser in arbitrary units, the dielectric being gaseous air. Remarks. 4-22 cm....eoe-.. Both inner and outer cones both at 20° C. 4-82 cm. .......- Outer cone at —185° C. Inner cone at 20° C., about. AaS emis 22. ee Outer cone at —185° C. Inner cone at —75° C. A 1 Osgia Lee ee Outer and inner cones both at —185° C. Hence it is clear that mere change of temperature of the metal work of the cone condenser does not affect its electrical capacity by more than 1, or perhaps 2, per cent., and any larger changes in capacity found on cooling must be due to a real change in the dielectric constant of any dielectric substituted for the air between the cones, and not to mere dimensional changes of the condenser itself produced by the cooling. Another matter to which our attention was directed was the of Organic Bodies, §c., at very Low Teniperatures. 257 question whether there was any sensible or serious lag in the tem- perature of the resistance thermometer behind the temperature of the dielectric. Our usual custom had been to immerse the condenser when prepared for use in liquid air, and cool down the whole mass to —185° C., and then raising it out of the liquid air to take tem- perature and capacity readings as it warmed up. The resistance thermometer was placed in the inner cone in a thin test-tube, and fixed in with fusible metal in the inner cone. We therefore tried one experiment with pure glycerine as dielectric, in which the electrical measurements were made as the condenser was slowly cooled, instead of being made as it slowly heated up. The process of cooling from —38° pt. to —201° pt. was allowed to occupy one hour and forty minutes. The values thus found for the dielectric constant for glycerine for this range of temperature were practically in agreement with those found when the condenser capacity was measured as it warmed up instead of cooled down. Table VI. I. Dielectric Constant of Pure Glycerine. Corrected galvanometer deflection when the condenser had air as dielectric = 3°92 cm. for 100 volts. Mean Temperature galvanometer in platinum _ deflection Dielectric degrees. in centimetres. constant. Observations. —38°0 2°85 50°5 Condenser charged with 1:434 —42°6 oe 52°8 volig. . Time = 3.0 P.M. — 46:0 3°00 53'3 —55'8 317 565 —64°8 3°03 03°38 = Time = 4.15 P.M. —98°8 2°70 3°95 Condenser charged with 17:0 volts. —119°5 2°20 S19 Time = 4.30 p.m. —201:0. 11°05 2°82 Condenser charged with 92°4 volts. Time = 4.45 P.M. The above values of the dielectric constant of glycerine are in very fair agreement with the values obtained by us during rising temperature.* We have also extended our former observations made with this cone condenser and the galvanometric method to certain other frozen electrolytes, and measured their dielectric constants at low temperatures. The results and substances are as follows :— * * Roy. Soe. Pree.,’ vol. 61, p. 324. 258 Profs. J. Dewar and J. A. Fleming. Dielectric Constants | II. Dielectric Constant of Dry Concentrated Sulphuric Acid (H280s). The corrected galvanometer deflection with air as dielectric was 3°92 cm. for 100 volts. The switch frequency was 120. Mean Temperature galvanometer in platinum deflection Dielectric degrees. incentimetres. constant. Observations. — 200°9 14°6 3°86 Condenser charged with 94:2 volts. —186-0 14:8 3°90 S ‘ - 5 —181°8 2°5 3°82 Condenser charged with 16:2 volts. —150°7 27 4°13 » ” ” 2» —129°8 2°95 4°33 Condenser charged with 16°9 volts. —110°0 6:0 7°25 Condenser charged with 17:0 volts. III. Dielectric Constant of Dry Concentrated Nitric Acid (NOH). Corrected galvanometer deflection with air as dielectric = 3°90 cm. for 100 volts. Mean Temperature galvanometer in platinum deflection - Dielectric degrees. in centimetres. constant. Observations. —201°7 9°05 2°36 Condenser charged with 94-2 volts. —182°7 ai, 2°46 ; —165°8 J2o7. 2°42 —1182 9°40 2°45 —129°7 10:00 2°62 Before carrying out these experiments with the concentrated nitric and sulphuric acids the condenser had been carefully re-gilt and a glass steady-pin substituted for the ebonite one. 1V. Dielectric Constant of Sodiwm Fluoride (NaF). (JO per cent. solution.) Corrected galvanometer deflection with air as dielectric = 4:04 cm. for 100 volts. Mean Temperature galvanometer in platinum deflection Dielectric degrees. incentimetres. constant. Observations. —199°3 9°3 2°33 Condenser charged with 942 volts. —1742 9°3 2°35 —149°2 10°35 2°55 —135'0 14°35 3°67 —125°0 19-6 5°05 —115°7 Ar 4, 7°07 Condenser charged with 15:2 volts. of Organic Bodies, §c., at very Low Temperatures. 259 The electrical resistance of the condenser at —200 pt. with frozen sodic fluoride as dielectric was 2000 megohms. , V. Dielectric Constant of Hydrosodic Fluoride (NaFHF). (16 per cent. solution.) Corrected galvanometer deflection with air as dielectric = 3:42 cm. | for 100 volts. Mean Temperature galvanometer in platinum _ deflection Dielectric degrees. in centimetres. constant. Observations. —201°8 8:87 2°28 —186:0 8°87 2°25 —169-0 8°95 2°30 —178'8 9:0 2°31 —148°8 On 2°49 —142°2 10°9 2°79 —131°5 15°4 402 VI. Dielectric Constant of Sodiwm$ Peroxide (Na,02). (5 per cent. solution.) Corrected galvanometer deflection with air as dielectric = 4°41 cm. for 100 volts. : Mean Temperature galvanometer in platinum deflection Dielectric degrees. in centimetres. constant. Observations. — 198-0 And 71 Condenser charged with 1°434 volts. —184°5 5° 87 VII. Dielectric Constant of Solution of Hydroxyl (H,O2). (20 per cent. solution by volume.) Mean Temperature galvanometer in platmum deflection Dielectric degrees. in centimetres. constant. Observations. —203°5 10°8 2°38 Condenser charged with 99:2 volis. Some experiments were then made by electrolysing freely certain electrolytes and freezing them with liquid air in the act of electro- lysis. The dielectric constants were then subsequently determined in the frozen state. 260 Profs. J. Dewar and J. A. Fleming. Dvtelectric Constants VIII. Dielectric Constants of Frozen Electrolytes, Electrolysed freely in the act of Freezing. Corrected galvanometer deflection with air as dielectric = 3°42 cm. for 100 volts. Mean Temperature galvanometer in platinum _ deflection Dielectric degrees. incentimetres. constant. Observations. (a) 5 per cent. aqueous solution of potassic hydrate electrolysed with 0°2 ampere and 8 volts. Hvolved gas = 5:1 c.c. — 200°0 3°85 71-4 Condenser charged with 1°434 volts. Electrical resistance of condenser ° 5000 megohms when frozen. (6) Water electrolysed with 1:0 ampere. —]98-2 8°70 2-47 Condenser charged with 98 volts. Hlectrical resistance of condenser 5000 megohms when frozen. (c) Water electrolysed with 2:1 amperes. —198-0 8°65 2°42 Condenser charged with 98-9 volts. It is evident that the action of electrolysis prior to, and during freezing has no sensible effect on the subsequently measured di- electric constant even though the surfaces of the cone condenser are strongly polarised in the act of freezing the liquid dielectric. We have then paid some attention to the possible cause of the high dielectric values of some substances at very low temperatures. It is clear from the above described experiments with the Nernst bridge that for organic bodies such as ethylic alcohol, amyl alcohol, ethylic ether, and glycerine we obtain practically the same dielectric values at the low temperature, both when they are measured by the galvano- metric method with a switch frequency of 120, and when measured by Nernst’s method with a frequency of 350, or about three times as great. On the other hand for certain other bodies such as the frozen dilute hydrates of potassium and rubidium, and the oxide of copper suspended in ice, the dielectric value at the low temperatures is much diminished by increasing the frequency. Subsequently to the completion of the experiments described in this paper the suggestion has been made by R. Abege* that the high di- electric values at low temperatures are due to polarization of the * “Wied. Ann.,’ vol. 62, . 249. of Organic Bodies, Sc. at very Low Temperatures. 261 electrodes of the condenser, and that the capacity measured is a polarisation capacity and not a true dielectric capacity, and he supports this contention by pointing out that whenever we have obtained a large dielectric value at the low temperature it has always been measured with an electromotive force of 1:434 volts, which is less than the ordinary full reverse electromotive force of polarization. There are, however, reasons for considering that this contention is not a valid one. In the first place we have always in all the measurements begun operations by testing the dielectric capacity of our condenser with an electromotive force of one Clark cell (= 1:434 volts) in order to see roughly whether the dielectric value was large or small. If it was a smail value we then gradually in- creased the electromotive force until a good readable galvanometer deflection was obtained. We never found that with an electromotive force of 1°454 volts, the dielectric constant of any substance was greater than with a much higher voltage. In the next place, in many cases we changed from a working electromotive force of about 20 volts to one of 1°434 volts, and we never found any marked discontinuity in the calculated value of the dielectric constant at that pomt. If our previous papers on this subject are examined the following instances may be found. Table Vil.—Measurement of various Dielectric Constants with different Electromotive Forces. Voltage with which con- Tempera- Dielectric. stant was Substance. ture. constant. measured. Reference. —_ — §89°4 27°6 19'S” > Roy: Soc: Proes’ an a 2Ti QOD 1-4 vol. 61, p. 318. —124°2 18-1 S 3 ae { fee Ge 4 p ibid, p. 321. Sodic chloride, —118°5 21°2 - 10 p.c. solution. . { 1140 22:3 1-4 \ nia, p. 307. Potassic chromate, f —112°4 16°9 a 0 30 p.c. solution. . ao 22:3 4, } Tid, Pp. 387. Cupric carbonate, f —124°7 16°5 182 10 p.c. suspension | —120°8 —_ 21°8 q } Bid, p- 390. Baric hydrate, f{—1780 23-9 V 5 5 p.c. suspension | —174°2 95:5 i } Bia, p- 372. Bismuth oxide, f—1292 199 17 : Ibid., p. 376 10 p.c. suspension | —127°3 QA, pg, Dine An examination of the above instances will show that if the electromotive force is changed from about 20 volts to.1°4 volts, it does 262 Profs. J. Dewar and J. A. Fleming. Dielectric Constants not make a greater difference in dielectric constant than can be properly ascribed to the accompanying change in temperature. [Again the following measurements were made at about the same voltage :— . Dielectric Charging Substance. Temperature. constant. voltage. Potassic Bicarbonate ... —166°5 2°80 19°8 Sodic Bicarbonate ..... —166°7 48°70 lite \ Ferric Chiorides::. 2... —133°8 4-23 19°8 Sodic Chloride......... —129°2 14°55 20°2 \ Cupric Carbonate ...... —132°7 3°42 18:2 Cupric Sulphate....... —133°2 16°40 19-7 lithic Hydrate ........ —198-0 3°23 19°8 Barco Wdelydrate ) as 2s... —196'8 20:10 19°5 It is unlikely that polarisation accounts for the differences between the dielectric constants of the above substances, taken pair and pair, when measured at nearly identical temperatures, and very nearly the same voltages.— November 8, 1897. | In order to settle the matter finally we propose, however, to re- measure a number of those substances which have shown high dielectric values at the low temperature when measured by the galvanometric method at a frequency of 120 but using in all cases an electromotive force of 100 volts. If under the larger electromotive force the dielectric values of — gome electrolytes still remain large, it will be difficult to ascribe this large value to polarization. The facts, however, admit of another interpretation. It is clear that the dielectric constants of some substances at low temperatures are vastly more susceptible to change of electromotive force fre- quency than is the case with others, and that the electric strain pro- duced by agiven electric stress varies in some cases enormously with the time of imposition of the stress but very little in others. Another argument against the view that these high dielectric values are due to polarisation, as ordinarily understood, is as follows: The results of most numerous experiments on water show that the dielectric constant at or near 0° C., is a number not far from 80. This value is obtained whether the electromotive force reversals are infinitely slow or whether they are very large. The resulls of the measurement of the electrical refractive index of water even with ether waves only 4 mm. in length, and, therefore, having a fre- quency of about 75 x 10", as given recently by Lampa,* indicate a number not far from 9°5 as the refractive index and hence give a, dielectric value of 90. There can be no question of polarisa- tion of electrodes in this last case. On the other hand an increase * ‘Wien. Ber.,’ Part 2a, p. 587, 1896 ; also p, 1049, 1897. of Organic Bodies, §c., at very Low Temperatures. 263 - in frequency which hardly affects the value of the dielectric con- stant of water is sufficient to greatly decrease that of ethylic alcohol, and at the same frequency the dielectric constant of ethylic alcohol was found by Lampa to have fallen to a value of 5. Hence ethylic alcohol is more sensitive to change of frequency than water. There is, therefore, no @ priori reason why we might not expect to find the same thing even in a much greater degree“in certain other bodies at lower temperatures, viz., a true high dielectric value for a certain frequency, but great sensitiveness to increase of frequency in such fashion that increase of frequency greatly reduces the dielectric value. At the present stage of the enquiry it seems undesirable to endeavour to regard the facts wholly from the point of view of one hypothesis as to the nature of elec- trolysis. We have again to mention with pleasure the assistance we con- tinue to receive from Mr. J. HK. Petavel in the observational part of these investigations. Note added December 8, 1897.. Received December 9, 1897. Since the above paper was printed we have repeated some of our . former experiments with the 5 per cent. solution of rubidic hydrate and the 5 per cent. solution of potassic hydrate, using the original method in which a condenser having the frozen electrolyte as dielectric is charged and discharged through a galvanometer, but employing much higher charging voltages. The object of these experiments was to apply a further test as to the validity of the contention put forward by R. Abegg, that we have obtained high dielectric values for certain of these frozen electrolytes in con- Sequence of haviug invariably used an electromotive force of 1°434 volts in the experiments with these particular substances. In order to be able to work with larger electromotive forces we arranged three galvanometers of the Holden d’Arsonval type otherwise exactly similar, except in having different sensibilities and resistances. One was the 500-ohm galvanometer used in all our previous condenser experiments, another was a 100-ohm coil galvanometer, and a third was a 4-ohm coil galvanometer. These were used at the same dis- tance (125 cm.) from the scale as formerly. An approximate test for the relative sensibility of these galvanometers was made by placing a Clark standard cell in series with each galvanometer through 100,000 ohms and noting the scale deflection produced. As the internal resistance of the galvanometers was at most only 4 per cent. of the total resistance these scale deflections may be considered to be approximately produced by the same current. The scale deflec- tions in centimetres were— 264 Profs. J. Dewar and J. A. Fleming. Dielectric Constants For the 500-ohm galvanometer.... 373 cm. Ke 100-ohm , ae Gry "ea bi 4-ohm : ae e's) aoe Hence, the sensibilities are in the ratios of these deflections, and the deflections of the 100-ohm galvancmeter must be multiplied by 5 5, and those of the 4-ohm galvanometer by 678, to reduce their scale readings to equivalent deflections in terms of the 500-ohm galvano- mMeueT, The 500-ohm galvanometer was then used with the condenser and vibrator, as described in one of our previous papers.* The con- denser having gaseous air as dielectric, the scale deflection for a frequency of 120 and an electromotive force of 97 volts was 3:2 cm. when corrected for capacity of leads. The condenser then had its dielectric space filled with the 5 per cent. solution of rubidic hydrate, and was frozen in liquid air. The dielectric constant was next measured, using an electromotive force of 17°8 volts and the 500-ohm galvanometer. The value of the dielec- tric constant found, when corrected for the capacity of the leads, was 65°6. The mean corrected scale deflection was 38°5 cm. for 17°8 _ volts. This, reduced to its equivalent for 97 volts, is 210 cm, and 2IO/3:2 =) 0d6: In the next place the same experiment was repeated employing the 100-chm galvanometer and an electromotive force of 79 volts. Applying the necessary corrections to the observed scale deflection of 23°5 cm., and reducing to the equivalent deflection on the 500-ohm galvanometer, gave 160 cm. as the value of the reduced deflection. Hence 160/3'°2 = 50 is the dielectric constant. The rather con- siderable difference between these values (65°6 and 50) is not a matter for surprise, having regard to the extreme steepness of the dielectric curve of the rubidic hydrate solution at about the temperature of liquid air. As we were merely desirous of determining whether a considerable increase of electromotive force wouid greatly diminish the large dielectric value, we did not trouble to put in operation the rather elaborate platinum thermometer arrangements for determining the exact temperature of the dielectric. A reference to the dielectric- temperature curve of rubidium hydratet will show that even one or two degrees of temperature change in the neighbourhood of —185° C., or —200° pt. makes a very considerable alteration in the dielectric value. The result, however, ascertained is that changing the electromotive force from 1°434 volts to 17°8 or 79 volts does not bring down the dielectric constant from a large value to a small one. In the same way the 5 per cent. solution of potassic hydrate was tested. * ‘Roy. Soc. Proc.,’ vol. 61, p. 800, 1897. + Ibid,, p. 378. of Organic Bodies, §¢., at very Low Temperatures. 265 Using the 500-ohm galvanometer and an electromotive force of 988 volts we found 153 as the dielectric constant of the frozen electrolyte at the temperature of liquid air. Employing the 4-ohm galvanometer and 79°5 volts we found 175 as the dielectric constant. All the above observations were taken at the temperature of liquid air and with an electromotive force frequency of 120. It is, therefore, quite clear that as far as these two frozen electrolytes are concerned, raising the charging voltage to a value far above that of the average electromotive force of polarisation does not bring down these abnormal values of the dielectric constant. On the other hand, a relatively small decrease in the temperature or dncrease in the frequency at low temperatures suffices to reduce the dieleciric value of these frozen hydrates very considerably.* We may, there- fore, say that the contention put forward by R. Abegg that the high dielectric values we have found for certain substances at the liquid air temperature are really polarisation capacities, does not seem to be borne out by the results of further experiment, and for the following reasons :— | (i) Because a very great increase in the charging electromotive force does not in any corresponding degree reduce the abnormally high dielectric values of certain frozen electro- lytes to much smaller values. (11) Because when in the course of observations to construct 2 temperature dielectric curve, the working electromotive force has been changed from a value below the counter- electromotive force of polarisation to a value far above it, there is no break or discontinuity in the curve of dielectric value. (iii) Because the great difference between quite similar electro- lytes, such as the 10 per cent. solution of potassic and sodic carbonates, in respect of dielectric constant at equally low temperatures and under equal charging electromotive forces is left unexplained. (iv) Because in the case of many substances, such as frozen ammonic hydrate, ice, and oxide of copper in suspension in ice, at very low temperatures, we find high dielectric values even though employing alternating currents of fairly high frequency (350 —). * The effect of increased frequency of electromotive force reversals in decreasing the dielectric constant is evidently dependent upon the temperature as welias on the physical state of the body. Inthe case of water an increase in the frequency from zero to 10° hardly affects the dielectric constant at all. In the case of ice at O°C. the same increase in frequency reduces the dielectric constant from 80 to between 2and3. In the case of ice at —50°C., as we have shown above, an increase in frequency from 120 to 350 reduces the dielectric constant from about 60 to about 3°6. (December 21, 1897.) VOL. LXII. U 266 Dielectric Constants at very Low Temperatures. (v) Becanse the high dielectric values of water and alcohol and other bodies at ordinary temperatures remain, even when the observations are taken with alternating electromotive forces of exceedingly high frequency and under conditions when there are no electrodes to polarise, as whcn the electric refractive index is measured with electromagnetic radiation. | We consider that the results of observations so far made are best expressed by merely considering the dielectric constant to be a function of the frequency and the temperature, and represented therefore by a dielectric surface, which surface has for some substances a region of abnormal dielectric ordinate. In all cases so far examined by us, lowering the temperature suffi- ciently acts in the same manner in reducing the dielectric constant as sufficiently increasing the frequency, and both actions reduce the abnormally large dielectric values of some substances to values more approximately equal to the square of the optical refractive index of the body. The question then to be considered is the physical reason for the high dielectric values for particular substances for certain ranges of electromotive force frequency and temperature. Whether this abnormal electric displacement is considered to be the result of a molecular strain superimposed on a true electric strain, or whether it is the beginnings of that molecular deformation which finally ends in chemical decomposition, remains to be seen. Having regard to the enormously high electrical resistivity which we have shown these frozen dielectrics to possess, 14 does not appear to us likely that polarisation in the sense of a deposition of ions on the electrodes can be invoked to explain the differences we have shown exist. Proceedings. 267 December 16, 1897, The LORD LISTER, F.R.C.S., D.C.L., President, in the Chair. Professor Gabriel Lippmann was admitted into the Society. A List of the Presents received was laid on the table, and thanks ordered for them, The following Papers were read :— : iY. alt EY. We. VII. “On a Method of determining the Reactions at the Points of Support of Continuous Beams.” By Gerorce Witson, M.Se., Demonstrator in Engineering in the Whitworth Laboratory of the Owens College, Manchester. Commu- nicated by Professor OsBporne Reynotps, F.R.S. **The Comparative Chemistry of the Suprarenal Capsules.” By B. Moors, M.A., Sharpey Research Scholar and Assistant in Physiology, University College, London, and SwaLe Vincent, M.B. (Lond.), British Medical Association Research Scholar. Communicated by Professor Scuirer, F.R.S. ‘Memoir on the Integration of Partial Differential Equations of the Second Order in Three Independent Variables, when an Intermediary Integral does not exist in general.” By A. R. Forsyru, F.R.S., Sadlerian Professor in the University of Cambridge. “On the Biology of Sterewm hirsutum, Fr. By H. Marsmaru Warp, D.Sc., F.R.S., Professor of Botany in the Univer- sity of Cambridge. . **An Examination into the Registered Speeds of American Trotting Horses, with Remarks on their value as Heredit- ary Data.” By Francis Gatton, D.C.L., F.R.S. “On the Thermal Conductivities of Single and Mixed Solids and Liquids, and their Variation with Temperature.” By Caries H. Les, D.Sc., Assistant Lecturer in Physics in the Owens College. Communicated by Professor Scuustsr, HBS. . “Cloudiness: Note on a Novel Case of Frequency.” By Karu Pearson, M.A., F.R.S., University College, London, U 2 268 Mr. G. Wilson. Ona Method of determining the VIII. “On the Occlusion of Hydrogen and Oxygen by Palladium.” By Lupwic Moyp, Ph.D., F.R.S., Witt1am Ramsay, Ph.D., LL.D. SED. ties nd JOHN Sarenps, D:Sc., Fiew: The Society adjourned over the Christmas Recess to Thursday, January 20, 1898. “On a Method of determining the Reactions at the Points of Support of Continuous Beams.” By GEORGE WILSON, M.Sc., Demonstrator in Engineering in the Whitworth Laboratory of the Owens College, Manchester. Communi- cated by Professor OSBORNE REYNOLDS, F.R.S. Received November 20,—Read December 16, 1897. The theory of continuous beams has been the subject of so much research in the past that further investigation would seem almost superfluous. In certain cases which occur in practice, however, the computations arising out of the existing methods become com- plicated and laborious, if not impossible to reduce, so that any solution which avoids these difficulties may be of sufficient value to warrant its publication. Mr. Heppel, in a paper read before this Society,” has traced the developments in the theory, culminating in the discovery of the “Theorem of Three Moments,’ by M. Bertot, in 1856, and independ- ently by M.M. Clapeyron and Bresse, in 1857. Previously to Clapeyron, Navier and other authors had sought the solution of the problem by obtaining the reactions at the various points of support of the beam; Clapeyron, however, first introduced the innovation of considering the bending moments at the points of support as she unknown quanti ties to be determined. M. Bresse, in his ‘Cours de Mécanique Appliquée,’ has discussed very fully the solutions of the various problems by this method, on the supposition of a constant moment of inflexibility of the sections of the beam both for the case of spans of arbitrary lengths and also for cases where the end spars are equal but of different length to the intermediate spans whose lengths are a!l supposed to be equal. Mr. Heppel, in the above mentioned paper, published solutions in which the spans were divided into two, three, four, or five equal parts throughout each of which the load and the cross section of the beam were supposed to remain constant, although varying from one division to another. Professors Perry and Ayrton} have dealt with the question of a * ©Roy. Soc. Proc.,’ vol. 18, p. 176. + Ibid., vol. 29, p. 493. Lteactions at the Points of Support of Continuous Beams. 269 yariable moment of inertia by obtaining the theorem of three moments in a slightly different form, the necessary summations for each span being performed graphically, whence on substitution in the original equations the bending moments can be obtained. The: author has reverted to the problem of finding the reactions at the - points of support and has based his method on a principle, definitely stated by Bresse,* and applied by him to the case of a uniform con- tinuous beam of two equal spans. The author claims that the method given below affords an easy and accurate solution for continuous beam problems, and especially those in which the moment of inertia is variable. It also permits of the variations in the stresses due to alterations in the levels of the supports being investigated. The principle may be reproduced as far as is oa as follows :— The displacement of any point by reason of the deformation of the beam is the resultant of the displacements which would be produced if one supposed all the external known forces to act separately and one after the other. : This being so, the continuous beam may be considered as a simple beam supported at each end and under the action of the given loading acting vertically downwards, and also under the action of the supporting forces at the intermediate piers acting vertically upwards. If the neutral fibre of the beam in the unloaded condition is assumed to be a straight. line, then the result of the action of these two distinct systems of loading is to make the final deflection of the neutral fibre at each of the intermediate points of support equal to zeYo. If the beam consist of » spans there will be n—1 intermediate supports, the upward pressure of each of which acting by itself would produce a definite deflection of the beam at any point: the sum of the separate deflections produced by these pressures at any one point of support must equal the deflection produced there by the given loading, and as each of the constituent deflections can be expressed in terms of the unknown concentrated load causing it, there will, therefore, be (n—1) equations each containing the reac- tions at each intermediate point of support, and as there are (n—1) reactions these equations are sufficient to determine their values. Let Ay A, A,.... An be the points of support. 2 = L. AvjA; = hs AvjA, = I, ie = le cove » * “Cours de Mécanique Appliquée,’ vol. 1, p. 187. 270 Mr. G. Wilson. On a Method of determining the Let m be the bending moment at any point in the mean fibre of the beam, due to the given system of loading. I = the moment of inertia of the section of the beam upon which m acts. HK = the modulus of elasticity. Let the origin be at Ay; the axis of 2 be A,Ai....A,; and the axis of y be perpendicular to that of 2 and positive downwards. Also let the suffixes 0,1, 2, 3,....refer to the corresponding points Ao, Ai, Ay, Ag,....so that m, is the bending moment at the first intermediate point of support, due to the given loading. Then from the equation E ary. wate de OI for the case of a beam supported at. each end, we obtain i! where T, is the tangent of the angle of inclination to the axis of z, of the tangent to the mean fibre at the origin— | (ny Cig be -|\ ae + ED, -where y, is the deflection at A, which would be produced were the beam only supported at each end and under the action of the given loading. Again let m', m", m'”, &c., be the bending moments at any point in the mean fibre due to the upward thrust of the reactions R,, R,, R;.... at the points A,, Ay, A,...., and T,, Ty, 0) game corre- sponding tangents at the origin. Ey = -||? de? +ET a, Ly 1 Then By! = -|| da? + ETL 0 (ely By," — -|| det ETO «0 7 hence finally since y=yty' t+y"+.... for the points Ay, A, As.... , we have aT oo hoy \ Br || A da? = (ntva.—|| ™ aot) +(B1y' || Ga aaa Ler a ape ly rly , lp dd . Er — || = da? = (wry2—| = an')+(B1".—{] = dz") “+, eeeg 0 \ «0 / 0 Reactions at the Points of Support of Continuous Beams. 271 and so on, there being as many equations as intermediate points of support. If the function m/I is assumed to represent the intensity of a new loading on the girder, it can easily be shown that the expression aa m —— ae I BT piciaas v¥/0 represents the value of the bending moment at the point Aj, due to this new loading, considering the beam as supported only at each end. Let this expression for the bending moment at A, be called Nj, and that at A, N., and similarly for A;..... Again, we may write the expression li, (ers1.—|| = dt equal to Ria’, ¢ J0 wi, being the bending moment at Aj, due to a new “m/I” loading obtained by assuming a unit force to act at Au. Similarly (onya—{f ia’) oe Esty"; Hence the equations become 3 N, = Ryn + Ren," + Ren" + «2... , er Rye Raia! A Retell eee a 5 Ng = Ryn; +.R.ng’ + Ryn + 2206 , &e., &e., , from which R,, R,, R;.... may be easily obtained when the con- stants have been determined. | Lord Rayleigh has shown, in his ‘ Theory of Sound’ (vol. 1, p. 69), that when a beam is loaded with a concentrated load at any point P,, and the load is transferred to a second point P2, then the deflection at P, when the load is at P, is equal to the deflection at P; when the load is at P., hence I aA = ue, <= 0, yay, == 1, 5 iC CC., thus reducing the number of constants to be found, or affording a check on the accuracy of the working. In the example appended 272 Mr. G. Wilson. Ona Method of determming the Nz’ = 22°47 by calculation. ty = 2258 4 Mean = 22°50 Hrror = 0°18 per cent. In practice it is usually very difficult to obtain the value of the {| 7 by integration unless I is assumed to be constant, or some 0 simple function of a. The following method will, however, give the values of N, x’, n.... required to any degree of accuracy. FEC. Let the diagram APCA (fig. 1), obtained by erecting ordinates proporticnal to the values of m/I at every point in the mean fibre AC, represent the new loading. The bending moment at any point B may be found very simply as follows :— If PN be any ordinate such tnat PN‘dz represents the load on any small element dx at N, then if PN be divided in Q so that QN/PQ = NO/NA, then QNdz represents that part of the reaction at A due to this load PNdz at N. Hence by taking a sufficient number of points a curve AQC can be drawn to !represent the reaction at A due to the load APC. Again, the load on AB may be replaced by a load at A having an equal moment about B, in a similar manner, either graphically as in the figure, which explains itself, or by calculation. Thus a second curve, ARB, is obtained. Then it is easily seen that the moment at B is the difference of the areas of the diagrams AQCA and ARBA, multiplied by AB, that is— = AB(area AQCBRA). This area may be obtained by a planimeter or by calculation, for Reactions at the Points of Support of Continuous Beams. 273 if all the ordinates to the curves are taken at equal intervals, the span being divided into an even number of the Jatter, the final diagram AQCBRA may be plotted and its area obtained by any planimeter, or by Simpson’s rule. . In the example appended the area has been found both ways. Elevation or Depression of Points of Support. Let 6 with the correct suffix represent the elevation of any point of support above the line A,A,, a depression being reckoned as negative. Then in the equations to find the reactions, since the final deflec- tion is not zero, we must write N,+ Eé, = Ryn’ + Ren" + eevee N,+Eé& = Rin!’ + Ren! + eoor « Some of the values of R,R,..... may be negative, in which case, if the beam is not to be fastened down at the supports, a fresh solu- tion must be sought by omitting one or more of the negative reactions, until the remaining ones become positive. The mean fibre of the beam has hitherto been assumed a straight line when under the action of no force. In certain girders this is not the case, but the above methods may be applied with sufficient accn- racy for practical purposes when the maximum distance of the external layer of the beam from the mean fibre is small compared with the original radius of curvature of the mean fibre. nm 1 a . For then r= (a) approximately, where R is the original radius of curvature of the mean fibre at any point and R’ its curvature after loading. Hence, if v = F(z) is the original equation to the mean fibre, and y =f (#) the equation after straining, then cae ay _ vv Bi da’ da? Gy av m dz? dz? °. KIL’ or which shows that the final deflection curve is the result of super- posing on the original curve of the mean fibre, the deflection curve obtained under the given loading, for a beam of the same cross 274 Mr. G. Wilson. On a Method of determining the — section at every point, but with a straight mean fibre, and hence the method is applicable. Appendix. To find the magnitude of the reactions at the points of support of a continuous girder of three spans (viz., 60 feet, 100 feet, and 40 feet respectively), and loaded with a uniform load of 3 tons per foot run, the moment of inertia of the cross section being 2100 at the commencement of the 60-foot span and increasing uniformly to 3300 at the commencement of the 100-foot span, and thence diminishing uniformly to 3000 at the beginning of the 40-foot span, and further diminishing uniformly to 2200 at the other end of the girder, the units taken being feet and tons. The variation of the moment of inertia is shown in fig.2. The curve A,BCA; in fig. 3 is the M/I curve for the beam resting on each end support, and loaded with 3 tons per foot run. Treating this as a new load and reducing the ordinates in the correct proportion, the curve A,DEA,; shows the amount of this load transmitted to A, to form the reaction there. FIG.2. r=3000 Reactions at the Points of Support of Continuous Beams. 270 Again the curves A,HA, and A,FA, show the loads acting at A,, which have their moments about A, and A, respectively, equal to the moments of the loads on A,A, and A,A, about A, and A, respec- tively. Hence N, = area A,DHA,A,HA, X 60 = 16330. N, = area A,DEA,;A,F A, xX 160 = 12084. Similarly the curve A,BC 7 ; On the Biology of Stereum hirsutum, Fr. 285 most interesting examples (from the point of view of application to mathematical physics) are the equations Wry = 0, Vu = —x’2, or, with the notation of the memoir, a+b+c=0, a+b+e= —«Kv; and the theory is applied to these equations in detail. Solutions, which are believed to be new, are obtained for both of them; each solution involves two explicit arbitrary functional forms, and the argument of each of these arbitrary functions itself involves an arbitrary element; but in each case the solution is not that of the widest possible generality which the equation is known to possess. To quote one result: a solution of the equation a+tbt+e=0 can be stated as follows :— Let p,q, r denote three arbitrary functions of w subject solely to the condition pt+_etr —09O : let w be determined as a function of a, y, z by means of the equation au = xp(u)+yq(u) +27(u), where a is a constant, and let v denote G(x) a— xp'(w) —yq'(u) —2r'(u) ” where G and H are distinct arbitrary functions: then v satisfies the equation H(u)+ atb+c=Vvu=v. “On the Biology of Stereum hirsutum, Fr.” By H. MARSHALL Warp, D.Sc., F.R.S., Professor of Botany in the University of Cambridge. Received November 23,—Read December 16, 1897. (Abstract.) The author has cultivated the mycelium of this fungus obtained from spores, on sterilised wood, and after several months the cultures doveloped yellow bosses which proved to be the hymenophores bearing the basidia. This fungus has not hitherto been made to produce x 2 286 Thermal Conductivities of Single and Mixed Solids, Ge. spores in cultures, and Basidiomycetes generally have rarely been made to do so. The actions of the mycelium on the wood of Msculus, Pinus, Quercus, and Sale are also examined, and this is, so far as known, the first time this has been done with pure cultures. Anatomical and histological details, with figures, are given in the complete paper. “ On the Thermal Conductivities of Smgle and Mixed Solids and Liquids, and their Variation with Temperature.” By CHARLES H. Less, D.Sc., Assistant Lecturer in Physics in the Owens College. Communicated by Professor SCHUSTER, F.R.S. Received November 30,—Read December 16, 1897. (Abstract.) These experiments were undertaken with a view to determining the effect of temperature on thermal conductivities, and the relation between the conductivity of a mixture and the conductivities of its constituents. The apparatus consisted of a number of flat circular copper discs, into each of which a thermo-junction was soldered. The substances to be experimented on were placed between these discs, heat was supplied to one of the dises at a measured rate, by passing an electric current through a coil in contact with it, and the differences of temperature between the discs were measured by balancing the thermo-electromotive forces produced, against the fall of potential down a wire. About thirty solids, liquids, substances near their melting points, and mixtures of liquids, were tested between temperatures of 15° and 50° C., and the following state- ments embody the results :— 1. Solids not very good conductors of heat in general decrease in conductivity with increase of temperature in the neighbour- hood of 40° C. Glass is an exception to this rule. 2. Liquids follow the same law in the neighbourhood of 30° C. 3. The conductivity of a substance does not invariably change abruptly at the melting point. 4. The thermal conductivity of a mixture lies between the con- ductivities of its constituents, but is not a linear function of its composition. 5. Mixtures of liquids decrease in conductivity with increase of temperature in the neighbourhood of 30° C., at about the same rate as their constituents. Cloudiness : Note on a Novel Case of Frequency. 287 * Cloudiness: Note on a Novel Case of Frequency.” By Karu Pearson, M.A., F.R.S., University College, London. Re- ceived December 1,—Read December 16, 1897. In a memoir on Skew Variation, contributed some time back to the ‘ Philosophical Transactions,* I pointed out (p. 364) that we might expect theoretically to occasionally find (J-shaped distribu- tions of frequency. I was unable at that time to refer to any case actually known to me except Mr. Francis Galton’s curve of “con- sumptivity.” The data in that case did not seem to me sufficiently definite to base any elaborate calculations upon them. Quite recently, in studying Hugo Meyer’s ‘Anleitung zur Bearbeitung meteorolo- gischer Beobachtungen fiir die Klimatologie,’ Berlin, 1891, I came across, on S. 108, the table for the frequency of various degrees of cloudiness for the decade 1876-85, at Breslau. Although the method used for determining the extent of cloudiness is not entirely satisfactory, and, as Herr Meyer remarks, the observer must have had some personal bias with regard to the grade 9, still the observations are so numerous, and so markedly (J-shaped, that I thought it well worth investigating how far my theory of skew variation would suffice to describe such a novel form of frequency. The observations are as follows :— Degrees of Cloudiness at Breslau, 1876—-1885. Derres ai.... 0 1 2 3 4, 5 6 7 8 9 10 Prequency.... 751 179 107 69 46 9 21 71 194 117 2089 The total number of days of observation is 3,653. Clearly no cloudiness and absolute cloudiness are both maxima while the mean cloudiness will not be very far removed from mini- mum frequency. The following data were obtained for the distribution by Miss Alice Lee, by the methods of the memoir referred to above :— Mean = 6'8292 Ei = 06112 faz = 18°2999 fp. = 1°7414 M3 = —61°2030 6+36,—2B2 = 4°3508 f4 = 583°1838 The theoretical curve is thus one of limited range. * Series A, vol. 186, 1895. 288 Prof. Karl Pearson. . 1 Proceeding we found e = 0'00699 r = 0°17958 Mm, = —0'8774 A, = 4°8109 Mz == —0'9480 ag == Y705 The negative values of m, and m, show us that the theoretical curve has changed from its usual form to a (J-shaped figure. The range given is b = a,+a, = 9'9814, instead of the actual 10. The distance d between mean and mode Ao —Q — 2 — 92-9022, rT Thus the start of the range is 4°8270—4°8109 = 0-016, instead of O, and it runs to 9°998, instead of 10. We conclude accordingly that if the range of possible cloudiness had been quite unknown a prvort, it would have been closely given by theory. The modal frequency y, was found to be 50°7505. Thus the theoretical equation to the frequency is— no —0°8774 Ay —0°9430 = 50°7505 ( 1-+—— yotaehs : d : ( ier = ( ar the origin being at 4°8270. The moda] value now corresponding to a minimum and not toa maximum as usual, the name “mode” ceases to be appropriate.* The observations and the above curve are given in the accompanying diagram, and it will be seen that there is a complete transforma- tion of the usnal frequency distribution to fit the altered state of affairs. With the asymptotic character of the curve, it is impossible to compare ordinates as giving the frequencies between 0 and 1, and 9 and 10. Accordingly the areas of the curve between : 0016 and 0O°'5, and between 9°5 and 9:998 were taken as the true measure of the frequencies of the degrees 0 and 10. These were obtained by means of the following formule :— Ay = M1 Neh aX a i _ Me Ne (% +1) [% \, (yob ) x (m (m+) = x( b a 4 2(3— ™m) \o ) A; = My Ng" we Lm fe \ | oy i an) * ck b ‘ees 2— Ne slg b )*3¢ 2 (3—m) (F . where m= —-My nz, = —Me, * The name antimode is now convenient. 289 a (36538 Days. x | >> S = D = [| & = S WD BS S| ~ D P= = S = D 8 — — 80° Between 14 and 13 mm. 45 » 103 and 9-9 mm. 60 3 76 and 7°2 mm. The minimum effective thickness is thus seen to uadergo a diminution with the increase of the angle of incidence. II. The influence of the Wave-length. In the following experiments I kept the angle of incidence con- stant, and varied the wave-length. JI used three different radiators, A, B, and C; of these A emitted the longest, and © the shortest waves. The following method of experimenting was adopted as offering some special advantages. If a cube of glass be interposed between the radiator and the receiver placed opposite to each other, the ea eee ee ee ee ee es Po ee Air-space on Total Reflection of Electric Radiation. 303 radiation striking one face perpendicularly would be transmitted across the opposite face without deviation and cause a response in the receiver. If the cube be now cut across a diagonal, two right- angled isosceles prisms will be obtained. If these two prisms were now separated slightly, keeping the two hypotenuses parallel, the incident radiation would be divided into two portions, of which one portion is transmitted, while the other portion is reflected by the air film in a direction (see fig. 1) at right angles to that of the Fie. 1.—Section of the two prisms. incident ray, the angle of incidence at the air-space being always 45°. The transmitted and the reflected portions would be comple- mentary to each other. When the receiver is placed opposite to the radiator, in the A position, the action on the receiver will be due to the transmitted portion; but when the receiver is placed at 90°, or in the B position, the action on the receiver will be due to the reflected portion. The advantage of this method is that the two observations for transmission and reflection can be successively taken in a very short time, during which the sensitiveness of the receiver is not likely to undergo any great change. In practice three readings are taken in succession, the first and the third being taken, say, for transmission and the second for reflection. I shall now give a general account of the results of the experi- ments. When the prisms are separated by a thickness of air-space greater than the minimum thickness for total reflection, the rays are wholly reflected, there being no response of the receiver in position A, but strong action in position B. As the thickness is gradually decreased below the critical thickness, the rays begin to be trans- mitted. The transmitted portion goes on increasing with the dimi- nution of the thickness of air-space, there being a corresponding diminution of the reflected component of the radiation. When the thickness of the air-space is reduced to about 0°3 mm., no reflected portion can be detected even when the receiver is made extremely sensitive. The reflected component is thus practically reduced to zero, the radiation being now entirely transmitted ; the two prisms, in spite of the breach due to the air-space, are electro-optically con- 304 Dr. J.C. Bose. On the Influence of the Thickness of i Fic. 2.—L is the lens to render the incident beam parallel; P, P’, are the right- angled isosceles prisms ; A and B are the two positions of the receiver. The receiver-tube is not shown in the diagram. tinuous. This is the case only when the two prisms are made of the same substance. If the second prism be made of sulphur, or of any other substance which has either a lower ora higher refractive index, there is always found a reflected portion even when the two prisms are in contact. Another interesting observation can be made by separating the prisms for total reflection. There would now be no transmitted por- tion. But if athin piece of cardboard or any other refracting sub- stance be now interposed in the air-space, a portion of the radiation will be found to be transmitted, and it will be found necessary to separate the prisms further to reduce the transmitted portion to Zero. Having given a general account of the experiments, I shall now describe the method of procedure. The radiator tube was provided with an ordinary lens whose focal distance for electric radiation is about 4 cm. The beam thus rendered approximately parallel fell perpendicularly on the face of the glass prism. The two prisms were made by cutting a cube of glass-—an ordinary paper weight— across a diagonal. The size of the cube was 4°5 cm. on each side.* One prism was fixed on the spectrometer circle; the other could be moved so as to vary the thickness of the interposed air-space between the two sections very gradually. The separation was simply effected by means of ordinary cards. The cards used were of uniform thickness, each card being 0°45 mm. in thickness. A certain number of cards were taken and placed between the prisms with their surfaces in contact with the hypotenuses. The cards were then carefully withdrawn, leaving the prisms separated by a thick- ness of air equal to the thickness of the given number of cards. It would, of course, be an improvement to have a micrometer screw by which the thickness may be gradually increased. * Larger prisms would have been preferred, had they been available. The prisms after cutting were found to be approximately isosceles, the angles being 90°, 46°, and 44°, Air-space on Total Reflection of Electric Radiation. 305 Observations were now taken to determine the minimum thickness of air for total reflection for different wave-lengths, the angle of inci- dence being in all cases kept at 45°. Three radiators, R,, R., Rs, were used. I have not yet made determinations of the lengths of wave emitted by these radiators, but it will be seen from the dimen- sions of the radiators that the waves emitted by R, are the longest and those emitted by R; the shortest. The oscillatory discharge in R, took place between two circular plates 1°2 em. in diameter and an interposed ball of platinum 0°97 cm. in diameter. The radiators were enclosed in a tube 3°8 cm. in diameter. In the radiator R,, the discharge took place between two beads of platinum and an interposed sphere the same asin R,. The distance between the sparking surfaces was 1°01 cm. In the radiator R;, sparking took place between two beads and an interposed sphere 0°61 cm. in diameter. The distance between the sparking surfaces was 076 cm. One prism was fixed on the spectrometer circle, and the other was at first placed somewhat apart from it; the distance was now gradually reduced till the air-space just ceased to reflect totally, when a small portion of radiation began to be transmitted. The beginning of transmission was detected by the receiver, which was placed in the A position. The detection of the beginning of trans- mission is, as has been said before, somewhat dependent on the sensitiveness of the receiver. Minimum thickness for total Radiator. Distance between spark- : reflection. ing surfaces in mm. | | | | | R, — Between 10°3 and 99 mm....... (a) R, 10°1 an €6 and, 7-2 mm... (b) R; 7°6 = 59 and 5'4mm....... (c) From the above results it is seen that the effective thickness of the totally reflecting air-space increases with the wave-length. If the wave-lengths are proportional to the distance between the sparking surfaces which give rise to the oscillatory discharge, the wave-lengths in (b) and (c) are in the ratio of 101: 76. This is not very different from the ratio of the corresponding minimum thicknesses of the totally reflecting air-space. III. On the Helation between the Reflected and the Transmitted Compo- nents of Radiation when the Thickness of Air-space undergoes Variation. _ In the general account of the experiments, I have said that as the _ thickness of air-space is gradually reduced the intensity of the 4 j 306 Dr. J.C. Bose. On the Influence of the Thickness of transmitted portion cf radiation is increased, while there is a corre- sponding diminution of the intensity of the reflected portion. This I have been able to verify qualitatively from numerous observations. But in making quantitative measurements many serious difficulties are encountered, owing to the difficulty of maintaining the intensity of radiation, as well as the sensitiveness of the receiver, absolutely constant. As regards the first, the intensity of the emitted radiation depends on the efficiency of the secondary spark, and the nature of the spark- ing surface. Keeping the primary current that flows through the Ruhmkorff coil constant, the efficacy of the secondary spark is very much affected by the manner in which the contact is broken in the primary circuit. If a vibrating interrupter is used, the break is apt to become irregular ; the torrent of the secondary sparks also spoils the sparking surface of the radiator. For merely qualitative experi- ments the use of a vibrating interrupter is not so very prejudicial, as along with the ineffective discharges there are present some which are oscillatory. But where successive discharges are to give rise to radiation of equal intensity, it becomes necessary to avoid all sources of uncertainty. For these reasons I prefer a single break for the production of a flash of radiation. With some practice it is possible to produce a number of breaks, each of which is effective. If the surface at the break is kept clean, and the break is properly effected, successive flashes of radiation up to a certain number are about equally intense. When the sparking has been taking place for too long a time, the surface no doubt undergoes a deterioration. But twenty or thirty successive sparks are equally efficacious when spark- ing takes place between platinum surfaces. The use of a single flash of radiation is preferable on another account. The receiver at each adjustment responds to the very first flash, but becomes less sensi- tive to the subsequent flashes. The conditions of the different experi- ments are maintained similar, when the action on the receiver is due to a single flash of radiation, instead of the accumulated effect of an unknown number of flashes. I give below the deflections of the galvanometer produced by four successive flashes of radiation. (GL) ibs geese : 115 divisions. CARPA EGE se 122 5 (Bye sieve evant 113 ‘5 (Ai & seei heueh onal 108 +5 When very careful adjustments are made, the successive deflec- tions are approximately equal. There are, however, occasional failures, owing either to the fault of the break, or loss of sensitive- ness of the receiver, naj Air-space on Total Reflection of Electric Radiation. 307 ‘More serious is the difficulty in connection with the receiver. With the improvements adopted there is no difficulty, under any circumstances, to make the receiver very highly sensitive; but it is extremely difficult to maintain the sensitiveness absolutely uniform. I have in my previous papers explained how the sensitiveness of the receiver depended on the pressure to which the spirals were subjected, and on the H.M.F. acting on the circuit ; and how the loss of sensitive- ness due to fatigue was counteracted by slightly increasing the H.M.F. For each receiver there is a certain pressure, and a corresponding E.M.F., at which for a given radiation the receiver is sensitive. Having obtained these conditions, the sensitiveness can be in- creased or decreased to almost any extent by a slight variation of either the pressure or the H.M.F. An increase of pressure pro- duced by the advance of the micrometer press screw through a fraction of a millimetre would sometimes double the sensitiveness ; similarly an increase of H.M.F. of even =i; volt increases the sen- sitiveness to a considerable extent. The nature of the difficulties in maintaining the sensitiveness of the receiver uniform will be understood from what has been said above. These difficulties are indeed great, and appear at first to be insuperable. But by very careful and tedious adjustments I was able on several occasions to obtain fairly satisfactory results, and was in hopes of ultimately obtaining symmetrical values from the galvanometer deflections. The setting-in of the rainy weather has unfortunately introduced other conditions unfavourable to the main- tenance of uniformity of the sensitiveness of the receiver. Owing to the excessive damp and heat the spirals get rusty in a short time, and variation in the sensibility is produced by the altered condition of the surface of the sensitive layer. The results of certain experiments I have carried out lead me to hope that this difficulty will, to a certain extent, be removed by covering the sensi- tive surface with a less oxidisable coating. The deflections produced in the galvanometer can only be taken approximately proportional to the intensity of the absorbed radia- tion. It would be better to observe the diminution of the resistance produced by the incident radiation. This may be done with the Henn of a differential galvanometer and a balancing resistance. Gis a high resistance differential galvanometer, with two sets of electrodes, A, B; C, D; one pair of electrodes is in series with the receiver, and the other with a resistance box. When the receiver is adjusted to respond to the electric radiation, a weak current flows through it. The same E.M.F. acts on both the circuits. The com- pensating current, produced by a proper adjustment of the resistance of the box, brings the spot of light back to zero. The resistance of the box is equal or proportional to the resistance of the receiver. 308 Dr. J. C. Bose. On the Influence of the Thickness of When radiation is absorbed by the receiver the resistance is decreased and this diminution of the resistance is found from the new balancing resistance. Fie. 3.—G, the differential galvanometer; R, the receiver; r, the resistance box. All observations agreed in showing that as the thickness of air-space was gradually decreased, the transmitted component was increased, with a corresponding decrease of the reflected portion. I give below two sets of observations, in which the receiver acted better than usual. The results are to be taken more as qualitative, as no reliance can be placed on the sensibility of the receiver being absolutely uniform. Radiator R,; distance between the sparking surfaces = 10°71 mm. Thickness of air- Galvanometer deflec- | Galvanometer deflec- Thickness space in terms of Sikes a tion due to the tion due to the number of cards. 1 reflected portion. transmitted portion. 1 | o- BO O or very slight. | Ageinst the stop. 2 | 0 | Stat Slight ae cl aa . bt 4, pes Tt baer 80 alae: eee 169 8 | 2 oa a 145 150 10 | 4 srl 150 120 12 | 5 *4 | 160 | 100 | 16 pes 2 | Against the stop 30 18 | Lt % 99 | 0 Air-space on Total Reflection of Electric Radiation. 309 lt is seen from the above, that as the thickness of the air-space was gradually increased, the reflected component increased, while the transmitted portion decreased. The minimum thickness for total reflection was found to be about 8 mm. When the thickness of air-space was reduced to about half this thickness (slightly less than half) the reflected and the transmitted portions seemed to be about equal. With the radiator R, the minimum thickness for total reflection was found to be about equal to the thickness of 22 cards (9°9 mm.). When the thickness of air-space was reduced to the thickness of 10 cards (4°5 mm.) the reflected and the transmitted portions seemed to be about equal. As two experiments immediately following each other are more likely to be comparable, the experiments were so arranged that the observation of deflection for transmission with a certain thickness of air followed the observation for reflection with a different thickness, the corresponding deflections being about equal. As stated above, the reflected and the transmitted portions were approximately equal when the thickness of air was equal to the thickness of 10 cards. Keeping 10 as the mean, pairs of readings were taken with different thicknesses. For.example, the refiection reading with a thickness of air equal tothe thickness of 4 cards was followed by taking a reading for transmission, with a thickness of air equal to the thickness of 16 cards; the deflections produced in the two cases were about equal, 1.e., sixty-six divisions of the scale. I append below a table showing the corresponding thicknesses of air (in terms of number of cards) which gave approximately equal deflections, the deflection in one case being due to the reflected com- ponent, and in the other case to the transmitted component. The receiver was made moderately sensitive, so that the deflections lay within the scale. S$ Thickness of air for Thickness of air Deflection produced. : reflection. for transmission. 4, io 66 6 14 | 70 | } 8 12 90 | 10 10 120 | When the thickness of air was reduced to 0°45 mm., a deflecticn of VOL. LXII. : Z 310 Dr. F. Galton. An Hramination into the two divisions was obtained for the reflection reading. From this an approximate idea of the intensity of the reflected component may be obtained. Half the total radiation gave a deflection of 120 divisions. The intensity of the reflected component, with a thickness of 045 mm., is ‘therefore 1/120th part of the total amount of incident radiation, on the assumption, which is only approximate, that the galvanometer deflections were symmetrical. When the thickness was reduced to 0°3 mm., no reflected component could be detected, though the receiver was made extremely sensitive. « An Examination into the Registered Speeds of American Trotting Horses, with Remarks on their value as Hereditary Data.” By Francis Gatton, D.C.L., F.R.S. Received November 29,— Read December 16, 1897. It is strange that the huge sums spent on the breeding of pedigree stock, whether of horses, cattle, or other animals, should not give rise to systematic publications of authentic records in a form suitable for scientific inquiry into the laws of heredity. An almost. solitary exception to the disregard, shown by breeders and owners, of exact measurements for publication in stud books, exists in the United States with respect to the measured speed of ‘‘trotters”’ and “ pacers” under defined conditions. The performance of 1 mile by a trotter, harnessed to a two-wheeled vehicle, carrying a weight of not less than 150 lbs. inclnsive of the driver, in 2 minutes 30 seconds qualifies him for entry in the Trotting Register, giving him, as it were, a pass-degree into a class of horses whose several utmost speeds or ‘‘records”’ are there published. To avoid prolixity I will not speak particularly of pacers (pace = amble), since what will be said of the trotters apples in general principle to them also. The great importance attached to high speed, and the watch- fulness of competitors, have resulted in evolving a method of timing trotters which is generally accepted as authoritative. The length of the track is scrupulously measured, and numerous other con- ditions are attended to, that shall ensure the record being correct, with an attempted exactitude to the nearest quarter of a second. A race against time, even if exact to the nearest quarter of a second, is by no means so close a measure of the speed of a horse relatively to his competitors, as the differential method of ordinary races. The speed of 1 mile m 2 30", or of 1760 yards m 150 seconds, is equivalent to about 12 yards in 1 second. Now, the length of a horse when extended at full trot is half as long again as his height at the withers—as I gather from the instantaneous photo- graphs of Muybridge—and consequently is hardly ever as much as registered Speeds of American Trotting Horses. dll 3 yards. Therefore at a 2’ 30" speed a horse travels through his whole length in a quarter of a second. In an ordinary English race a winner by half a length gains a notable victory, while a neck or even a head in advance is sufficient to establish his priority. ‘There- fore the record of the speed of a horse to the nearest quarter of a second is by no means an absurd refinement. It is, of course, very dificult under the exciting circumstances of a race to measure time with such precision as that. I tested the value of these entries as follows :—If quarter seconds were noted with exactness the entries of 0, +, 4, and 2? would be approximately equal in number; they would also be equal if they were set down at random without bias, but if there be a bias towards favourite numbers its effects would be apparent. I extracted a few hundred entries, and found the rela- tive frequency of the 0, +, 4, and ? to be almost exactly as 1, 3, 2, and 1. Consequently the } is on the average three times as great a favourite as either the 0 or the 3, and the 4 is twice as mucha 2 favourite as they are. It is evident that the = seconds are not 4 strictly trustworthy, but it may well be urged that their entry is preferable to their total disregard. , I was informed: that a trifling laxity was tolerated when a horse had just but only just failed to qualify, an allowance of ; of a second in his favour being commonly made. So that a speed of 2’ 30}” would usually be reckoned as 2’ 30’. I shall return to this point further on. The system of timing and of registering records began more than fifty years ago, and was developed and improved by degrees. In 1892 a considerable change was made in the conditions by the introduction of bicycle wheels with pneumatic tyres, which produced a gain of speed, the amount of which is much discussed, but which a prevalent opinion rates at 5 seconds in the mile. Thenceforward the records are comparable on nearly equal terms. All trotting performances up to the 2’ 30" standard are registered in the large and closely printed volumes of ‘ Wallace’s Year Book,’ published under the authority of the American Trotting Association. Vols. 8—12 refer to the years 1892-6, and it is from_the entries in these that the following remarks are based. The object of my inquiry was to test the suitability of these trotting (and pacing) records for investigations into the laws of heredity. Their trustworthiness was of course one point to be ascer- tained, another was to obtain a just notion of the proper principle on which marks for speed should be awarded, as, for instance, in the following example :—Suppose a particular ancestor, whom we will eall A, of a certain horse has a record of 2’ 30”,and that another ancestor in the same degree, whom we will cail B, has a record of 2' 10", how are their joint influences to be estimated ? Will it be the z 2 312 Dr. F. Galton. An Kxamination into the same on the average as that of two horses each having the speed of 2’ 20", or will it be something altogether different? In short, is the arithmetical the most appropriate mean or not? It would be a strong presumption in the affirmative, if the relative frequency of the various speeds should correspond approximately with those determined by the normal law of frequency, because if they do so they would fall into line with numerous anthropometric and other measures which have been often discussed, and which, when treated by methods in which the arithmetic mean was employed, have yielded results that accord with observed facts. Whether the speeds do or do not occur with the normal frequency had therefore to be ascertained. So my inguiry had two objects: first, did the run of the observations suggest a tolerably smooth curve? Secondly, was that curve a tolerable approach to the curve of normal fre- quency ? The investigation was troublesome and tedious. It was necessary to pick out from a large collection the names of those stallions, geldings, and mares (all three being equally efficient trotters), whose records had been made in the year under consideration, and who also had arrived at maturity, that is, who were not less than five years old, and therefore had had time to show their fuil powers. Had younger horses been included, the frequency of the slower records would have been: much increased. Assisted by a friend, the appropriate entries were underlined in the printed volumes, then one of us read them out, and the other ticked them down in the appropriate column of a page ruled for the purpose. Finally the marks in each column were counted. In this way 5705 extracts were made from the entries for the years 1892-96; they were not subsequently verified, so some few omissions are probable. Anyhow they form a fair and large sample, and are quite sufficient for the present purpose. The discussion of this material resulted in rather bulky tables, which it is needless to reproduce here, because their contents are given in an adequate and much simpler manner by the accompanying diagrams. The successive columns in the table are represented in the diagrams by imaginary columns that stand on corresponding bases. They run as follows :—The first column, counting from the left, contains the percentage value of all observations recorded as 2’ 29°0", 292”, 294", or 293"; that is of all under 30 down to 29 inclusive (the minutes being here omitted for brevity). The second column referred to 28:0", 281’, 284", and 282”, and soon with the rest. Con- sequently the dot in the diagram which indicates the percentage number of observations, according to the side scale, stands in the middle of its own imaginary column. For example, that of the 2’ 28" set stands vertically above the point that lies half way J registered Speeds of American Trotting Horses. 313 between 28 and 29 on the scale along the base. The dots are con- nected by thin lines to show the trace or curve of the observations. The smooth curves are those of normal frequency, calculated from the values of the mean (M) and of the probable error (P.E.), which are given in the diagrams. Freguency. FreEQuesiCy. 50 Sec 20 /0 Freguericy. 30 Secs 20 /O Leaving aside for the moment the strange pinnacle that rises on the extreme left of every diagram, we see that the traces of the observations run very roughly, but not intolerably so. In each diagram they seem to be disposed about a fundamentally smooth eurve. Considering the smallness of the interval, namely, only 1 second, that separates the observations assigned to each pair of suc- cessive columns, together with the experience derived from other kinds of statistical curves, it seems to me that the run of the obser- 314 Dr. F. Galton. An Examination into the 20 o esl FFEGUESICY. S ie Nf vatious is good enough to certify their general trustworthiness. As regards the pinnacle it is a different matter, and is one which when beginning work, as I did, on the 1892 entries only, was very perplexing. However, by persevering with the other years it became increasingly plain that the pinnacle was a false maximum; in 1896 it was certain that the true maximum lay well within the portion of the curve included in the diagram. The explanation of the pinnacle then became obvious; it was that the tolerance granted to those horses who failed by only a little to qualify themselves, was extended considerably beyond the quarter second for which I was prepared.* The cases of 2' 30:0" were few; they do not appear in the diagram, but their addition would be quite insufficient to remove the difficulty. If the pinnacle were distributed among two adjacent columns outside and to the left of the diagram it would smooth away the incongruity, so I suspect that cases of “under 2' 32’ and down to 2’ 30”” are habitually rated at a trifle less than 2' 30’. Consequently I had no hesitation in wholly disregarding the entries that helped to make the pinnacle, namely, the whole of those contained in the first column to the left in every one of the diagrams. The course thereupon became clear and straightforward. When fixing upon the mean for each year, I was somewhat biassed by the entries in the adjacent years; simi- larly as to the probable error. Now that the curves are drawn I see that somewhat better fits might have been made, but they are close enough to show the existence of a fair amount of correspondence between the observed values and those calculated according to the law of normal frequency. It is near enough to remove hesitation in working with the arithmetic mean. * (Jan, 20.—I have since learnt that the conditions of timing are too rigorous to justify this inference; also that the very numerous efforts simply to secure a standard record, and thenceforward to cease training, may be a chief cause of the pinnacle. | ~ registered Speeds of American Trotting Horses. 315 I now come to the fundamental purpose of this memoir, which is to point out the existence in the registers of the American Trotting Association, of a store of material most valuable to inquirers into the laws of heredity, which accumulates and increases in value year by year. Unfortunately it lies buried to a hopeless depth, partly because the published part of the registers refers only to standard trotters. It appears to be buried simply through the omis- sion of having its importance insisted on. The published volumes of the ‘ Trotting Register’ contain numerous elaborate tables, but lacks one that should include the names and pedigrees of those horses concerning whose antecedents enough is known to make their pedigrees serviceable to investigators. ; It is hardly worth while to discuss hereditary influence on speed, in the case of any horses, unless the records of at least their sires and of their dams, and those of each of their four grandparents, as well as their own record, are all known. [Even in this case (according, at least, to my own theory) one quarter of the hereditary influences are unknown and have to be inferred. It is practically impossible to make an adequate collection of the names of horses who fulfil the above conditions out of the entries in the ‘Trotting Register,’ each search requiring many cross references and occupying a long time, while the number of futile searches before attaining a success is great. On the other hand, the breeders and possessors of these notably bred horses must be familiar with the required facts, and would assuredly be delighted to have them known. There need, therefore, be little difficulty in obtaining materials for the much desired table. In the meantime I am sending circulars to the chief breeders in America in hopes of making a start. The great need for genealogical data of an exact numerical kind, by those who prosecute inquiries into the laws of heredity, is the justification that 1 offer for submitting these remarks to the Royal Society. 316 Prof. W. Ramsay and Mr. M. W. Travers. January 20, 1898. The LORD LISTER, F.R.C.S., D.C.L., President, in the Chair. The Right Hon. Sir Nathaniel Lindley, Master of the Rolls, one of Her Majesty’s Most Honourable Privy Council, was balloted for and elected a Fellow of the Society. A List of the Presents received was laid on the table, and thanks ordered for them, The following Papers were read :— I. “The Relations between Marine Animal and Vegetable Life.” By H. M. Vernon, M.A., M.B. Communicated by Professor J. Burpon Sanperson, F.R.S. Il. “The Homogeneity of Helium.” By Wiutiam Ramsay, Ph.D., LL.D., Se.D., F.R.S., and Morris W. Travers, B.Sc. III. “ Fergusonite, an Endothermic Mineral.” By Witiiam Ramsay, Ph.D., LL.D., Sc.D., F.R.S., and Morris W. Travers, B.Sc. TV. “On the Modification of the Spectra of Iron and other Sub- stances, radiating in a strong Magnetic Field.” By Tomas Preston, M.A. Communicated by Professor G. F. Fr1z- GERALD, F.R.S. “The Homogeneity of Helium.” By WILLIAM Ramsay, Ph.D., LL.D., Sc.D., F.R.S., and Morris W. TRAVERS, B.Sc. Re- ceived December 2, 1897,—Read January 29, 1898. About a year ago, a paper by Dr. Norman Collie and one of the authors (W. R.) was published, bearing the title ‘*‘ The Homogeneity of Helium and of Argon.” In that paper*™ various reasons were adduced to show why an attempt to determine whether or no argon and helium are homogeneous was worth making. The results of the experiments at that time indicated that while it did not appear possible to separate argon into two portions of different densities, the case was different with helium. Samples were obtained after * ‘Roy. Soc. Proc.,’ vol. 60, p. 206. The Homogeneity of Helium. 317 repeated diffusion which possessed respectively diffusion rates corre- sponding to the densities 2°133 and 1°874. It was there pointed out that these densities are not correct (although their ratio is probably not wrong), owing to the curious fact that the rate of diffusion of helium is too rapid for its density, 7z.e., it does not follow Graham’s law of the inverse square root of the densities. These samples of gas also differed in refractivity, and the difference was approxi- mately proportional to the difference in density. Towards the end of the paper, the conjecture was hazarded that it was not beyond the bounds of possibility that the systematic dif- fusion of what we are accustomed to regard as a homogeneous gas, for example, nitrogen, might conceivably sift light molecules from heavy molecules. It is true that the fineness of the lines of the spectrum would offer an argument in favour of the uniformity of molecular weight; but still it is never advisable to assume any physical theory without submitting it to rigorous proof. And it was thought possible that the fact al diffusion to which helium had ' been subjected might have had the result of effecting such a separa- tion; a separation, not of chemical species, but of molecular magni- ae The other and more ordinary explanation of the splitting of helium into fractions of different density is that helium must be regarded as a mixture of two gases, one lighter than the other. Since the publication of the paper mentioned, Dr. A. Hagenbach has confirmed the possibility of separating helium into portions of two densities by diffusion; and the differences in density were practically the same as those observed in the laboratory of Univer- sity College.* These experiments were made with somewhat over 200 c.c. of gas; but it was decided to make experiments of a similar kind, on a much larger quantity of helium. An apparatus was therefore constructed, similar in principle to the one previously employed, but on a much larger scale. The main features are shown in the illustration on p. 208 of the paper previously alluded to; but on account of the large amount of gas - diffused, it was not practicable to collect it in tubes. Instead, there- fore, of the bent tube EN of the former apparatus, the tube con- nected with the stopcock H was continued horizontally, and by means of six vertical branches it communicated with six gas reser- voirs, each furnished with a two-way stopcock. It was possible with this means to cause gas from any one of the reservoirs to enter the diffusion apparatus A. In order to be able to collect the gas in any desired reservoir, the delivery tube of the Topler pump F delivered gas into a jar somewhat similar to that shown at J, but provided with a vertical branch, which was bent horizontally some * © Wied. Ann.,’ vol. 60, p. 124. 318 Prof. W. Ramsay and Mr. M. W. Travers. distance up, and lay parallel to the previously mentioned horizontal tube. It, too, had six vertical branches, each of which communi- cated with the other limb of the two-way stopcock of each reservoir. By raising the reservoir of the Tépler pump and expelling gas into the collecting tube J, the gas could be transferred to any one of the reservoirs. The accompanying diagram makes it clear how the apparatus was set up. The actual method of conducting a diffusion was as follows :— Reservoir I was raised until the mercury in the diffusion jar A stood at the level of the dotted line. The clip L was then closed, and the stopcocks C and D opened. The Tépler pump was then worked until all gas was removed from A; the gas, if air (as at the commencement of the whole series of operations), being allowed to escape by moving the collecting jar J, so that it no longer covered the end of the exit tube of the Topler pump. Stopcocks C and D were next closed, and stopcocks HE and 6a opened, so that the gas from 6 entered the diffusion vessel A. By raising the reservoir belong- ing to 6, all gas was expelled through H into A, clip L being opened meanwhile. Reservoir 6 was now full of mercury, and all gas was in A. Stopceck C was then opened, and the gas in A diffused through the pipe stem B (closed at one end by means of an oxy- hydrogen blowpipe) into the pump. This diffusion proceeded until half the gas in A had passed into the pump reservoir F. Stopcock C was then closed, and the Tépler was worked, the diffused gas being delivered into J. Stopcock 6a was then opened, and the reservoir of The Homogeneity of Helium. a19 6 lowered, so that the gas in J passed into 6. This stopcock was then shut. The contents of 5 were then transferred in a similar manner into A, and one-third of the gas was diffused into the pump. It was collected as before in 6. The diffusion jar A now contained as much gas as had been present in 5. The contents of 4 were next added; half of this was removed by diffusion and transferred to 5. The contents of 3 were added; half was diffused and transferred to 4. The contents of 2 were added; half was diffused and collected in 3. And, lastly, the contenis of 1 were added, and the half dif- fused collected in 2. Stopcock D was then opened, and the mercury in the diffusion jar A allowed to run up to the dotted line; the clip L was closed. All gas was pumped out of A and collected in 1; this constituted one complete round. As it was not possible to empty the tube issuing from J com- pletely of gas by lowering the reservoir of 1, and as, if not emptied, the heavy gas would have contaminated the light gas from 6 during the next round, the following method was made use of. The gas from 6 was transferred to the empty reservoir A; and then, by lowering the reservoir of 6, mercury rose in the tube issuing from J, and expelled all the heavy gas in the connecting tubes into 6. The clip K was then closed, and by opening the stopcocks la and 6a, so that communication took place between jars 1 and 6, the small quan- tity of gas in 6 was transferred tol. The apparatus was now ready for a second round. The Fractional Diffusion of Atr. In order to test the working of the apparatus, a set of diffusions was carried out with air. After four rounds, comprising twenty-four diffusions, the light portion contained 17°37 per cent. of oxygen and the heavy portion 22°03. A fairly rapid separation was thus being effected, considering the closeness of the densities of nitrogen and oxygen. | The Fractional Diffusion of Nitrogen. A similar set of experiments was carried out with nitrogen, pre- pared by the action of solutions of ammonium chloride on sodium nitrite, in presence of copper sulphate. The gas was dried and passed over red-hot iron prepared by reduction of ferric oxide in order to remove any oxygen or to decompose any oxides of nitrogen which might be present. After thirty rounds, involving 180 opera- tions, the “light” portion of the nitrogen, after purification by circulation over copper oxide, had not altered in density. It must therefore be concluded that nitrogen is homogeneous as regards the relative density of its individual molecules. 320 Prof. W. Ramsay and Mr. M. W. Travers. The Fractional Diffusion of Helium. The first sample of helium employed was prepared from samar- skite and cleveite. After seventeen rounds, involving 102 operations, the diffusion rates of the lighter and heavier portions were measured. The first gave a density, calculated from this rate, of 1807, and the second of 2°128. The same gas was re-diffused until in all thirty rounds had been carried through, involving 180 operations, The light fraction now showed the density (measured diffusion rate against hydrogen) 1°816, and the heavy fraction 2124. These gases were then circulated; the diffusion rate of the lighter portion pointed to a density of 1811; the heavier gas was diffused into three por- tions, of which the more rapidly diffusing had a “ diffusion density ” of 1:906, and the less rapidly diffusing of 2022. The lightest gas of all (diffusion density = 1:311) was weighed, and had a “real” density (O = 16) of 2:021; the mixture of the heavy products gave the real density, 2:153. The refractivity of the heavy portion, measured against helium from cleveite, undiffused, yet purified from all removable gases, which had the density (weighed) 2°076, was 1078, the refractivity of the undiffused gas being taken as unity. A fresh quantity of helium was next prepared from cleveite, and the former diffused samples were stored in tube-reservoirs for future use. The new helium was washed with caustic soda, but not other- wise purified. This gas was now put through fifteen rounds, com- prising ninety operations, and the light portion in jar 6 was purified by circulation over magnesium and copper oxide. Its refractivity was 0°9752 of that of the uncirculated helinm. Its density by weighing was 1979. Owing to the cracking of the glass apparatus the main bulk of the specimen was lost. It may be here interesting to chronicle that the remaining portion was inhaled through the nose and mouth ; it possessed neither smell nor taste. The contents of No. 5 were therefore purified and weighed; its density was 2049. The contents of No. 1 were also purified by circulation, and had a gravimetric density of 2°245. It lost on circulation a considerable amount of nitrogen which was estimated as ammonia by treat- ment of the magnesium containing nitride with water. As we are certain that there was no entry of air in preparing the gas, the 34 ¢.c. of nitrogen must have been evolved from the mineral. It may have been occluded on the surface of the powdered mineral ; it need not be remarked that before heating the mineral a nearly perfect vacuum was made in the tube, and that there was no leakage during the operations. We have previously found traces of nitrogen in gas prepared from cleveite; but not all specimens give off that gas. Supposing, however, to take the worst view, the nitrogen had been The Homogeneity of Helium. d21 derived from leakage of air, it would correspond to only 0°3 c.c. of argon. The contents of jar No. 2 were also parified and weighed. During the purification hardly a trace of nitrogen was removed. The density was 2°209. We have thus :— Jar No. 1 contains gas of density...... 2°245 o 2 53 BI i ae a 2°209 _ 6 ai AMULET oh, 4 tae oes The light gas which had previously been stored in tubes was now mixed with the light gas from the second set of diffusions, and the mixture was re-diffused fifteen times, involving ninety operations. The density of the lightest portion of this helium was determined by weighing and found to be 1°988. The helium had, therefore, not been made sensibly lighter by re-diffusion. The mean of the two determinations may be taken as the true density of pure helium; it is 1:98. The refractivity of this sample measured against hydrogen and multiplied by the ratio between hydrogen and air, viz., 0°4564, gives 0°1238. This specimen of light helium of density 1:988 was placed in one of the refractivity tubes, and the lightest helium of the former preparation (density = 1:979) in the other. They had the same refractivity (1000 to 1004). The contents of No. 1, obtained from the mixture of light gases had the density 2°030, showing that only a little heavier material had been withdrawn. The lighter fractions of helium were then sealed up in glass reser- voirs and stored. The heavier portions were placed in the diffusion apparatus and submitted to methodical diffusion. | After fifteen rounds (ninety operations) the heaviest fraction had density 2°275, the lightest 2:08. The refractivity of the heaviest gas was next determined and found to be 0'1327. This gas examined in a Plicker’s tube showed brilliantly pure helium lines, but along with these the reds and green groups of argon. Calculating from the density of this gas it should contain 1°63 per cent. of argon according to the equation 1:96lz+20y = 2°275. Calculating from the refractivity the percentage of argon should be 1:05, from the equation 1:245a + 0:9596y = 13°33. A mixture of 99 per cent. of the purest helium and 1 per cent. of argon was made, and it showed the argon spectrum with about the same or with somewhat less intensity than the heaviest gas. Finally, the heavy gas was dif- fused to the last dregs, so that only about 0°5 c.c. remained undif- fused ; and this small residue, transferred to a Pliicker tube, showed the argon spectrum with only a trace of the spectrum of helium. The yellow line and the bright green line were visible, but feeble. This spectrum was compared with that of a mixture of argon with a trace of helium, and nearly the same appearance was to be seen. With 322 Prof. W. Ramsay and Mr. M. W. Travers. the jar in parallel and a spark gap interposed the blue spectrum of argon was equally distinct in both tubes; and, more important still, there was no trace of any unknown line. It appears, therefore, that helium contains no unknown gas, nor is it possible to separate it by diffusion into any two kinds of gas; all that can be said is that most minerals which evolve helium on heating also evolve argon in small quantity. This accounts for the difference in density observed in different samples of helium; and in one instance, viz., malacone, the amount of argon evolved on heating the mineral, though small, was much in excess of the helium, so far as could be judged by the spectrum. In the light of the experiments of which an account has here been given, it is necessary to reconsider the deduction drawn by Professors Runge and Paschen from the complex nature of the spectrum of helium as regards its complex nature. Sir Norman Lockyer has already pronounced in favour of the supposition that helium is a mixture, chiefly on the ground that in the spectra of certain stars some, but not all, of the helium lines are observable. It appears to us that this may well be accounted for by the hypothesis that the differences of temperature and pressure in the stars might produce variations in the spectrum of helium. If a jar and spark gap be interposed while observing the visible spectrum of helium, a pro- found alteration is to be noticed. The yellow line D; is to be seen near the electrodes, and is faint in the capillary portion of the tube, and one of the red lines disappears. The change is not as remark- able as in the case of argon, but is quite distinct and characteristic. Then, as before remarked, the green line becomes relatively stronger at low pressures, so that the hght evolved in the tube is no longer the usual brilliant yellow, but dull greenish-purple. Is it not likely that the conditions obtaining in the stars may account for the absence of some of the lines ordinarily visible P If the hypothesis of Runge and Paschen is correct, then the two gases to which they attribute the complex spectrum of helium must have nearly the same density. It has already been shown that by means of the apparatus used for the fractional diffusion of gases it is possible to effect a fair separation of the constituents of air after a few rounds. If the supposed constituents of helium differ in den- sity in as high a proportionas 14 to 16, it is certain that some separa- tion would have been effected. As there has been no such separation, the legitimate inference is that the density of the two supposed con- stituents does not differ by so great an amount, or that their exist- ence is imaginary. Itappears to us that too little is known regarding the nature of the vibrations which cause spectra to make it legitimate to theorise on the subject. It is surely conceivable that an atom may possess such a structure as to render it capable of propagating two The Homogeneity of Helium. 323 different sets of vibrations, each'complete in itself, and each resem- bling the other in general form. Yet it must be acknowledged that our experiments have not disproved the existence of two gases in helium of approximately the same density ; in fact it may be contended that helium is a pair of elements lke nickei and cobalt. We are disappointed in the result of this long research, because we had thought it not improbable that an element of density 10 and atomic weight 20 might prove to be the cause of the fact that different samples of helium possess different densities, according to the mineral from which they are extracted, and also of the separation of helium into portions of different densities by diffusion. We still regard it as by no means improbable that further research will lead to the dis- covery of the ‘‘missing’”’ element, and this. appears. to be a fitting opportunity of stating our reasons for the belief. _ The difference between the atomic weights of helium and argon is 40—4 = 36. Now, there are several cases of such a difference. If we compare the groups of which the first members are fluorine, oxygen, nitrogen, carbon, boron, beryllium, and lithium, we obtain the following table :— PUAOLINGajx.., 415 6): 050: eee ORGY Ate jmig kM. Ss LO ces Cil@rinies dx. shsares 35°5 a: AL \ay MAUNA GTI 25950") 9.0; 27:0 oa Manganese........ 59°0 PCMAG ship aps: ose fer 44] Sayeed ss. ea. ee 16:0 DiEIo AMAT dd ghee aac 971 Prlpuar.......... 32'0 “a Magnesium “S.).'.’. 24°3 ae Chromium,....... 52°3 ; Calerumy * LA, 40°] | Bi eogem) 66. %,..-,,-,. 140 ih nee re) Phosphorus ...... gah OO tsi vane cio ene Ch 51-4 POUASSUMIM oo. og soi ao Carben WN e.. 13:0 Belmamarnt ou 2am) 4:0 so 28:3 ie Daceracgps ts otal 29-9 160 Titanium ........ 48:1 Co ee 40-0° 70? The elements helium and argon have been given a provisional place. The differences between the extreme members of these small groups are given in the short table which follows :—- Manganese—Fluorine.... 86:0 Chromium—Oxygen .... 363 Vanadium—Nitrogen .... 37:°4 Titanium—Carbon ...... 36:1 Scandiuam—Boron ..,... 331 Calcium—Beryllium .... 31:0 Potassium—hithium .... 331 Argon—Helium The difference between the atomic weights of argon and helium, it 324 The Homogeneity of Helium. will be seen, is not far removed from those of the other pairs of elements. It appears, therefore, not improbable that there should be an element with atomic weight 20, resembling both argon and helium in its properties. Yet it is not so certain that the middle element should resemble argon and helium, for in the table given it is seen that there are several examples of elements with a middle place which do not resemble those at the extremes. The question is perhaps best left open. It will be remembered that the gases evolved from a great many minerals and mineral waters have been examined, and that in many cases they have been found to contain helium and argon. In no instance up to the present has any sample of the gases evolved on heating in vacnum been found to show unknown spectrum lines. The amount of argon, as proved by the account which we have just given of our experiments, is very small, and in the case of the gas from cleveite investigated by Langlet it is probable that argon was almost completely absent, for it possessed the density 2. In mala- cone, on the contrary, argon is present in larger amount than helium, although neither gas is obtainable from it in large quantity. It appears to us not beycend the limit of probability that in some as yet uninvestigated mineral the middle member of the helium group may be discovered. When it is considered that germanium, an element which has been recognised only in one of the rarest of minerals, argyrodite, is the middle element of the trio, silicon, germanium, and tin, of which the first and last members are common, it is surely not unreasonable to hope that the middle member of the helium trio may ultimately be found. The amount of helium in fergusonite, one of the minerals which yields it in fair quantity, is only 33 parts by weight in 100,000 of the mineral, and it is not improbable that some other mineral may contain the missing gas in still more minute pro- portion. If, however, it is accompanied in its still undiscovered sources by argon and helium, it will probably be a work of extreme difficulty to effect its separation from these gases. Addendum.—Since this paper was written, Professors Runge and Paschen, in a communication to the British Association in August of this year, have withdrawn their contention that helium is a mixture, or, perhaps more correctly stated, they now ascribe to helium the same complexity as that of oxygen, the spectrum of which may also be arranged in two series, each consisting of three sets of lines. As oxygen has not yet proved to be complex, the surmise that helium is cowplex therefore falls to the ground. Fergusonite, an Endothermic Mineral. d25 «F ergusonite, an Endothermic Mineral.” By WiuLIAM Ramsay, Ph.D., LL.D., Se.D., F.R.S., and Morris W. TRAvERS, B.Sc. Received December 15, 1897,—Read January 20, 1898. The mineral fereusonite, discovered by Hartwall, occurs in felspar and mica deposits, in the same manner as most of the rare Norwegian minerals, such as euxenite, orthite, samarskite, &c. The position in which such minerals are found, embedded in masses of felspar, or encrusted with mica, leaves the question of their origin an open one. Whether they are deposited in the felspar by water, or whether they are contemporaneous with the felspar, is a matter-of speculation. Fergusonite is a black lustrous mineral, not unlike obsidian in out- ward appearance, but of considerably higher density. Seen under the microscope, even with the highest power, there is absolutely no sign of crystailine structure, though in thin slices the substance is translucent, and transmits yellow-brown light. It is, however, macrocrystalline, occurring in quadratic sphenoids. It is quite homogeneous, and displays no sign of cavities. Like similar mine- rals, it contains helium, which is expeiled on the application of heat. But this mineral presents a peculiarity, which has led us to publish this note. When heated to a temperature not exceeding 500° or 600°, it suddenly becomes incandescent, and evolves much of its helium; while its density decreases. The analysis of the mineral was kindly undertaken by Miss Emily Aston, to whom we desire to express our indebtedness. The mineral has been previously analysed by Hartwall, its discoverer, and by Weber, and, for the sake of comparison, we quote the earlier analyses* :— Composition of Fergusonite. Miss Aston. Hartwall. Weber. Oxides of niobium and tantalum.... 40°95 A775 48°84 Oxides of yttrium, erbium, &....... 31°09 41°91 30°61 Rexides Or CEMUM, WC... 66 os 20 bc te oe BES Bey 4683 3:05 Ream GIOXIUG,. ses « «0's 0 5 3°36 — 0°95 0°35 irate GriOxide: ¢.6i6...6 381 — —— — LY Meaty GiOXIde 6. be. e ee ee ba e's 4°56 de ES 2 i a — 3°02 6°93 Bm ro ae solo dics 0's anes erate 1°42 — — LS Gea ee rr I-95 -— =— re oh Piel cece eee x00 —— 0:31 33 [Seal Que) oo is 0:16 — — Rep re oe c oye”. 5b erwiscne i ahs — 1:00 O85 TSEROR VOC reo cies 5 oe 2 os 0) ois 0-12 — -- 100°89 99°62 99°46 * Rammelsberg’s ‘ Mineralchemie,’ p. 401. VOL. LXIL. ye 326 ~ Prof. W. Ramsay and Mr. W. M. Travers. The oxides of niobium and tantalum were converted into double fluorides of these metals with potassium fluoride ; and on examination of the crystals under the microscope, they were seen to be almost entirely of one form. They were easily soluble in water, and, from previous experience with these compounds, we were able to recognise them as potassium niobium oxy-fluoride. There appears to be hardly any tantalo-fluoride present in the possible mixture. The uranium dioxide was estimated by heating the mineral with dilute sulphuric acid in a sealed tube, and titrating the dioxide with potas- slum permanganate. The trioxide was calculated by difference from the total uranium. The cerium metals were separated, as usual, by means of a saturated solution of potassium sulphate. lt is thus seen that fergusonite is mainly a niobate of yttrium, containing oxides of uranium, but in no great quantity. The gases evolved by the incandescence of nearly 5 grams (4°852) of the mineral, heated in a vacuous tube, had the following compo- sition :— Per gram of Total gas. mineral. Per cent. CG: Cr: JELSW LION ene Slane 0°24, 1-080 75°50 Hydrogen | ...%.% s p0S8 0-078 5°47 Carbon dioxide .... 1:19 0°245 17:14 INTEROP EN. cs elie) < ic 0-15 0:027 1:88 6°94 1-430 32°99 The remaining mineral was mixed with hydrogen potassium sulphate, and heated to redness. More gas was evolved; oxygen, resulting from the decomposition of the sulphuric anhydride, was present in considerable quantity. The sulphur dioxide and the carbonic anhydride were removed by passing the gases through soda- lime, before it entered the pump; hence they do not appear in the analysis. Per gram of Total gas. mineral, Per cent. C.c. c.c. Helium .. 24.2 So¢gen = ao 0733 60°3 INGirOvens cs ec ce 0-42 0-088 73 Oxyren\. lisse 1:87 0°394 32°4 O77 1-215 100°0 The mineral taken weighed 4°744 grams. The density was determined before and after heating. Great care was taken to make sure of the absence of air-bells, by warming the powdered mineral under water in a vacuum, before weighing it. Fergusonite, an Endothermic Minerai. O20 Density before heating...... 5°619 ¥ after ear ave dane CO ae It is thus seen that the mineral loses density on incandescence. The amount of heat lost by this curious mineral in parting with its helium was determined. The plan of operation was to burn in oxygen a known weight of hydrogen, ascertained by measuring it, under a small platinum crucible, in a calorimeter. The rise of tem- perature was noted. This operation was repeated several times, so as to standardise the calorimeter. Some grams of mineral were then placed in the crucible, and the operation was repeated ; the heat evolved by the incandescing mineral added itself to that from the burning hydrogen, and the rise of temperature was greater. Knowing the heat of combustion of hydrogen, a simple calculation gave the heat evolved by the exothermic change in the mineral. The actual data are as follows :— ue aie FEL T¥; Rise of temperature per gram of J C00 SOA 14°65° 14°68° 14°47° 14°56° Additional rise for 60595 grams ~ Le) ee areata pin of afoe, ate, — (aoa per cram Additional rise for 40830 erams Tie a Be ac ah'esw ona? eva’ eo, == Osea. Mi Mean rise per gram hydrogen .... 14°59° Mean rise per gram mineral ...... 0°345° Heat of combustion of 1 gram hy- lil DL 24h ee «» 934200 calories. Heat of decomposition of 1 gram MCP cia Mela asin 4 ae 20 + 0,0 , 809 99 In these experiments, a correction was of course introduced for the change of temperature of the calorimeter during the experiment, due to the temperature of the surrounding air being higher or lower than that of the calorimeter. The percentage of helium in the mineral, by weight, is 0:0194, evolved on incandescence, and on further heating, 00132; the total percentage is 9°0326. Dr. Shields was so kind as to determine the specific heat of fergusonite. A Bunsen’s calorimeter, in thorough working order, was used. The data are :— ere OF mineral. oi: i oe. ee es cite a cag oro 8:789 grams. Temperature before introducing into calorimeter.. 17:3° ©. Deflection (1 mm. = 0:001053 ‘K) LYSIS UREN ta a oct 154'4 mm. Mean specific heat between 0° and 17°3° ........ 0°1069 328 -Fergusonite, an Endotherme Mineral. Various questions are raised by the behaviour of this interesting mineral. Its evolution of heat, accompanying its parting with helium, suggest the idea thai it is a true endothermic compound of | helium. Had its density, as is the case with alumina, and with other oxides which rise spontaneously in temperature when heated, in- creased instead of decreasing, the cvolution of heat might justly have been ascribed to polymerisation. But an evolution of heat, accom- panied by a fall in density, leads to the conjecture that the loss of energy is the result of the loss of helium; and that, conversely, the formation of the compound must have been concurrent with a gain of energy. That the helium is actually in combination, and not retained in pores in the mineral, is evinced by there being no pores in which the helium might be imprisoned. Surface-absorption is equally out of the question, for the mineral is compact. The only remaining possibility is that the helium is in chemical combination. And if this is true, then the compound must be an endothermic one. The question next arises, with what constituent of the mineral is the helium in combination? This question cannot at present be answered. All that can be said is that the amount of helium does not appear to depend on the total percentage of uranium, although minerals containing uranium usually (probably always) contain this element. Even in English pitchblende there was found a trace of helium. And in malacone, a mineral containing no uranium, a trace of helium was found; also in a specimen of meteoric iron. The presence of niobic and tantalic anhydrides, and of the yttrium group | of elements, is also favourable to its presence. But the proportion between the weight of the helium and that of the other elements present makes any calculation of the atomic relations between the helium and the other elements out of the question. There is one other substance at least which decreases in density while it evolves heat; that substance is water, in changing into ice. The effect of compressing ice is to lower its melting point, and at the same time to reduce its heat of fusion. At a sufficiently high pressure there would be a continuous transition from ice to water, no heat change taking place during the transition. Matters would be in a similar condition to those which accompany the change of a liquid into gas at the critical temperature; the smallest alteration of tem- perature would be enough to bring about the change. Inspeculating on the origin of such a remarkable compound, is it not allowable to guess that it represents a condition of our earth realised only before solidification had set in? That these minerals, contaimuing the rare elements, represent a portion of the interior of our planet; and that under the enormous pressure obtaining at the centre, combination with helium was an exothermic event; and that such compounds, having by some unexplained accident come to the surface of the Proceedings and List of Papers read. 329 globe, where they are no longer exposed to such pressure, they have, in consequence of the change, become endothermic? The frequency of the helium spectrum in the stars, and its presence in the sun, makes it less improbable that some such explanation may lie not far from the truth. 3 There are at least two other minerals, gadolinite and eschinite, which exhibit endothermic properties. But these minerals, instead of decreasing in density on ignition, increase. The following table shows the gases evolved when they are heated, their densities before and after heating, and the loss of weight which they suffer :— Gases evolved. c.c. per gram. Density. a === oN Kas x = Loss of Hy. CO. CO,. He. Before. After. weight. Gadolinite .... 0°700 OO11 1060 none 4289 4371 0-82 Aischinite .... 0458 none 0215 0-243 4685 4°793 1-018 It is to be noticed that only the eschinite contains helium, and that in very small quantity. The fact that these minerals increase in density, and that only one yields helium, places them in a different class from fergusonite. Moreover, the rise of temperature is not to be compared to that seen with fergusonite, for the glow is barely visible. January 27, 1898. The LORD LISTER, F.R.C.S., D.C.L., President, in the Chair. A List of the Presents received was laid on the table, and thanks ordered for them. The Right Hon. Sir Herbert Eustace Maxwell, a member of Her Majesty’s Most Honourable Privy Council, was balloted for and elected a Fellow of the Society. The following Papers were read :— I. “ Mathematical Contributions to the Theory of Evolution. On the Law of Ancestral Heredity.” By Karu Pearson, M.A., F.R.S., University College, London. i. “On the Zoological Evidence for the former Connection of Lake Tanganyika with the Sea.” By J. E.S. Moors. Communi- cated by Professor LanKEsTER, F.R.S. 330 Proceedings and List of Papers read. Til. “The Kelvin Quadrant Hlectrometer as a Wattmeter and Volt- meter.” By H. Witson. Communicated by Dr. Hopxtyson, E.R.S. IV. “The Magnetic Properties of almost Pure Iron.” By HE. Winson. Communicated by Dr. Hopkinson, F.R.S. February 3, 1898. STR JOHN EVANS, K.C.B., Treasurer and Vice-President, in the ‘Chair. A list of the Presents received was laid on the table, and thanks ordered for them. The Right Hon. Sir Nathaniel Lindley, Master of the Rolls, was admitted into the Society. The following Papers were read :— I. “On the Intimate Structure of Crystals. Part I. Crystals of the Cubic System with Cubic Cleavage. Haloid Salts of the Alkalis.” By W. J. Sonnas, D.Sc., F.R.S. II. “On the Intimate Structure of Crystals. Part Il. Crystals of the Cubic System with Cubic Cleavage. Haloid Compounds of Silver.” By W. J. Scnnas, D.Se., FBS: IJ. “ Comparison of Oxygen with the extra Lines in the Spectra of the Helium Stars, 8 Crucis, &c.; also Summary of the Spectra of Southern Stars to the 35 Magnitude and their Distribution.” By Frank McCunay, F.RB.S. IV. ‘Researches in Vortex Motion. Part III. On Spiral or Gyro- static Vortex Aggregates.” By W. M. Hicks, F.R.S. V. “The Pharmacology of Aconitine, Diacetyl-Aconitine, Benza- conine, and Aconine, considered in relation to their Chemical Constitution.” By Joun THropore Casu, M.D., F.R.S., and WyrynpHam R. Dunstan, M.A., F.R.S. VI. ‘Note on the Experimental Junction of the Vagus with the Cells of the Superior Cervical Ganghon.” By J.N. Laneuey, D3¢,, 0.18.5. Junction of Vagus with Superior Cervical Ganglion. 381 “ Note on the Experimental Junction of the Vagus Nerve with the Cells of the Superior Cervical Ganglion.” By J. N. LANGLEY, D.Sc., F.R.S., Fellow of Trinity College, Cam- bridge. Received January 26,—Read February 3, 1898. Two experiments were made on cats. The central end of the vagus, cut a little below the larynx, was turned forward and jomed to the peripheral end of the cervical sympathetic. ‘The object of the experiments was to see whether the vagus nerve fibres are capable of forming connexions with any of the structures with which the spinal nerve fibres of the cervical sympathetic are normally con- nected. The results seem to me to be esi as regards ‘this point. The time allowed for regeneration was in one case 73 days, and in the other 123 days. At the end of these periods anesthetics were again given, and the nerves stimulated. Stimulation of the sympathetic in the lower region of the neck, i.e., of its central end, gave no effect of any kind. Hence the central end of the sympathetic had formed no functional connexions with the peripheral end. Stimulation ofthe sympathetic a little below the superior cervical ganglion caused reflex effects of the kind caused by vagus stimula- tion. These reflexes were obvious in the case in which 123 days had been allowed for regeneration, less clear in that in which seventy- three days only had been allowed. They ceased on section of the vagus close to the ganglion of the trunk. Thus afferent fibres of the vagus had grown outwards amongst, or joined with, the fibres of the peripheral end of the sympathetic. The stimulation also caused all the effects normally produced by stimulation of the cervical sympathetic, so that, although the central end of the sympathetic had not joined the polled end, the peripheral end had acquired more or less completely its normal function. Stimulation of the vagus a little below the ganglion of the trunk —the nerve being cut centrally of the’ point stimulated—caused dilation of the pupil, retraction of the nictitating membrane, con- traction of the arteries of the ear, erection of the hairs of the face secretion of the sub- maxillary gland, and the other effects soe le caused by stimulating the cervical sympathetic. After injection of nicotine no effect was obtained by stimulating the nerves centrally of the superior cervical ganglion; the usual effects following when the ganglion itself was stimulated. Hence, piesa: fibres of the vagus had either grown along the VOL. LXII. 2B 302 Prof. W. M. Hicks. peripheral end of the cervical sympathetic, and formed nerve- endings around the cells of the superior cervical ganglion, or they had united directly with the sympathetic fibres. That the former had taken place I infer from the fact that the regenerated nerve contained rmedullated fibres larger than those proper to the sympa- thetic. T conclude from the experiments that there is no essential differ- ence between the efferent “‘ visceral” or “involuntary” nerve fibres, whether they leave the central nervous system by way of the cranial nerves, by way of the sacral nerves, or by way of the spinal nerves to the sympathetic system. All of these fibres I take to be pre-gangtionic fibres. And I think that any pre-ganglionie fibre is capable, in proper conditions, of becoming connected with any nerve cell with which a pre-ganglionic fibre is normally con- nected; although apparently this connexion does not take place with equal readiness in all cases. On the whole it appears to me that the functions exercised both by pre-ganglionic and by post- ganglionic fibres depend less upon physiological differences than upon the connexions which they have an opportunity of making during the development of the nervous system and of the other tissues of the body. A fuller account of the observations will be published in the ‘Journal of Physiology,’ after some further experiments have been made. “ Researches in hs Motion. Part Ill. On Spiral or Gyro- static Vortex Aggregates.” By W. M. Hicks, F.R.S. Received January 12,—Read February 3, 1898. (Abstract. ) A portion of the communication (Sect. II) extends the theory of the simple spherical vortex discovered by Hill. The chief part (Sects. I and IIT) refers, however, to a kind of gyrostatic aggregate. The investigation has brought to light an entirely new system of spiral vortices. To give an idea of the species of motion considered, take the case of motion of an infinitely long cylindrical vortex of sectional radius a. The velocity perpendicular to the axis inside the vortex will be of the form v = f(r) where f(o) = 0. Outside it will be given by v= Va/r where V = f(a). ; We may, however, have a motion in which the fluid moves parallel to the axis inside the cylinder with rest outside. The velocity will be of the form u = F(r) inside, where F(a) = 0, and zero outside. Both f(r) and F(r) are arbitrary functions subject only to the condi- tions f(0) = 0 and F(a) = 0. Putting aside for the present the Researches in Vortex Motion. dae question of the stability of these simple motions or of their resultant, it is clear that if we superpose the two we get another state of steady motion in which we have vortex filaments in tbe shape of helices lying on concentric cylindric surfaces. The problem to be considered is whether it is possible to conceive a similar superposi- tion of two motions in the case of any vortex aggregate whose motions are symmetric about an axis. The general conditions for the existence of such systems are determined in Sect. I, and are worked out in more detail for a particular case of spherical aggregate in Sect. III. It is found that the motion in meridian planes is determined from a certain function yin the usual manner. The velocity along a parallel of latitude is given by v = f(¥)p where p is the distance of the point from the straight or polar axis. The function y satisfies an equation of the form (when expressed in polar co-ordinates) | 2 ee ee ee an ar or i@ =r ae °° where F and f are both functions of y. The case F uniform, and f «wis treated more fully. If f = \W/a where a is the radius of the aggregate, : hs po a Nevs= JA. { Je (~) rho Jay} sin’@. The most striking and remarkable fact brought out is that as Xx increases we get a periodic system of families of aggregates. The members of each family differ from one another in the number of layers and equatorial axes they possess. According to the number of jndependent axes they are called singlets, doublets, triplets, &., in contradistinction to more or less fortuitous or arbitrary compounds of the former, which are considered later and called monads, dyads, triads, &c. Of these families two are investigated more in detail than the others, both because they are specially interesting in their properties and because they serve as limiting cases between the different series. In one family (the A, family) all the members remain at rest in the surrounding fluid. In the other (the i, family) a distinguishing feature, common to all the members, is that the stream lines and the vortex lines are coincident. The parameter \ gives the total angular pitch of the stream lines on the outer current sheet, although in aggregates with more than one equatorial axis these lines are not one continuous line. The first agegregates—with \<5°7637 (the first A, value)—behave abnormally. Beyond these we get successive series, im one set of which the velocity of translation is in the same direction as the polar motion of the central nucleus, in the alternate set the velocity is opposite, and 282 334 Prof. W. M. Hicks. the aggregate regredes in the fluid as com — with its central aggregate. Suppose the attempt made to obtain sets of aggregates with greater and greater angular pitch. It will be found that as the external pitch of the stream lines increases the equatorial axis contracts and the surface velocity diminishes. On the outer layers (ring-shaped) the spiral pitch is chiefly produced on the inner side facing the polar axis until oni the boundary itself the stream lines he along meridians and the twist is altogether on the polar axis. The pitch can be increased up to a certain limit. As this is done the stream lines and the vortex lines fold up towards one another, coincide at a certain pitch, and exchange sides. When au external angular pitch of about 330° is attained it is impossible to go further if a simple aggregate is desired. Ifa higher pitch is desired the aggregate splits into two concentric portions—an inner spherical portion and an outer shell. The central nucleus is similar to those just described—it produces a part of the required pitch. The outer layer has spirals with the same direction of twist which complete the balance of the pitch. In these, however, the motion is in the opposite direction. With increasing pitch this layer becomes thicker and its equatorial axis contracts relatively to the mid point of the shell until another limitis reached ; the stream and vortex lines again fold together, cross, and expand as this second limit is reached. If a larger pitch still is desired there must be a third layer, and so en. The first coincidence of stream and vortex lines takes place for an aggregate whose pitch is 257° 27'. Whenever a maximum pitch is attained the aggregate is at rest inthe fluid. This is first attained when the pitch is 330°14'. Beyond this there are two equatorial axes. For a pitch 442°37’ the stream and vortex lines again coin- cide, the internal nucleus gives 257° 27’ of the pitch, and the outer shell the remainder, and so on. At the end of the paper a theory of compound aggregates is developed. It is not worked oat in detail in the present communica- tion, but the conditions are determined for dyad compounds, whilst a similar theory holds for triad and higher ones. Hach element of a poly-ad may consist of singlets, doublets, &c. The equations of condition allow three quantities arbitrary—as for instance ratio of volumes, ratio of primary cyclic constants, and ratio of secondary cyclic constzats. The full development of this theory is, however, left for a future communication. If we take any particular spherical aggregate with given \ and primary cyclic constant uw, the energy is determinate. We may, however, alter the energy. If it be increased the spherical form begins to open out into a ring form whose shape and properties have Researches in Vortex Motion. 335 not yet been investigated. If the energy be increased sufficiently the aperture becomes large compared with.the thickness of the cere and approximate calculation is applicable. The differential equa- tion for y in terms of toroidal co-ordinates is Sine: but the = development is left for a future occasion. In the paper itself the problem is treated purely as a question of _ hydrodynamics, and the results simply as the properties of certain possible fluid motions. It may not, however, be out of place here to offer a few remarks, of a more speculative kind, on the bearing of the results on physical theories. In the first place, gyrostatic motion of the kind here considered. is not confined to aggregates, which are symmetrical about an axis. Although the theory is very complicated, it is easy to see that they must exist. In the address to Section A, at the Ipswich meeting of the British Association, a vortex cell theory of the ether was indi- cated. The ether consisted of closely packed elements, each element being a vortex aggregate. To fix ideas, the case of elements of the shape of a rectangular box was taken, although this particular shape is not essential. The vortical motion there considered was not gyrostatic, but it is clear that a gyrostatic modification is possible. The primary rotations must be arranged in opposite directions in alternate cells. This is, however, not necessarily the case with the _ secondary gyrostatic motion. ‘They may either be or not be in the same direction, although conditions of stability might decide this question. If the common direction is not a necessity, it is easy to conceive that certain operations on boundaries immersed in the ether might make them so, and in this way produce the same effect as vortex filaments stretching between them. Such a theory would not necessitate return vortex filaments such as are required in any theory which attempts to explain electrical actions: by such filaments. It is very conceivable that they would produce the stresses along and perpendicular to tubes of force which are required in an electric field. lfacell, such as that of the \, aggregates in this paper, were possible, the necessity that the primary rotations should he alter- nately directed would not exist, at least so far as continuity of motion had to decide. In the second place, does the new theory throw any light on:a vortex atom theory of matter? In this respect two remarks should be made. The first is, that if vortex atoms are realities the exact quantitative theory developed in this paper cannot accord with actual facts, because it is developed with reference to a surrounding irrota- tional ether, which cannot be the case in nature. Nevertheless, many of the general properties would doubtless be siciileie’ and pessibly the same for ager egates of the d, family. — The second remark. is, that the results of the paper eee only to 335 Prof. W. M. Hicks. spherical aggregates, that is, all the various elements are compared, not when their energies are in thermal equilibrium, but in the arti- ficial association such that the energy of each particular element; is that which is necessary to give it a spherical shape. Nevertheless, it is possible to get general ideas. The most striking one is the fact of the periodic property of the atoms. The J, curve, for instance, or the curve in the figure which shows how the translation velocity alters with increasing pitch of spiral, irresistibly suggests curves con- nected with the physical properties of the elements. The abnormal commencement, the regular ascending and descending series suggest the connection at once, and open a vista of possibilities before unsuspected. For the reasons mentioned above, it would be waste ot time to look as yet for any definite information. Before that can be done we must know more about the conditions of stability, and the behaviour of such aggregates when their energy changes. It is — hardly fitting perhaps to indulge in wild speculations in these pages. In doing so, however, I hope they will be taken for what they are intended, merely as vague intimations of possibilities. Let us then take the well known curve showing how the fusibili- ties of the elements alter periodically with the atomic weights. In a solid body the atoms or molecules can have very little trans- latory motion. They will therefore take such forms, or their energy will be such as to make this translation small. Now take a spherical ageregate. If it has a large translation velocity its energy must be diminished to render this Jess—it will take a more elongated form with a small velocity of translation. In order, therefore, to fuse the substance more energy must be put into it. Its temperature of fusion is higher. In other words, it is natural to suppose that those atoms, which when in the spherical form have a high velocity, will possess high fusing points, and so on. Without criticising this argument too closely, let us make the assumption that it is so, and see what it leads to. Now look at the figure which gives the relation between the velocity of translation (ordinates) to the spiral pitch (abscisse). We are at once struck with the fact that we have aggregates with large maximum velocity followed with sets of small maximum velocity (in the opposite direction). This is one of the most remarkable features of the fusibility curve. Suppose that the curves march together: this supposition enables us to locate roughly the regions in which the elements lie, omitting the early ones as abnormal. If this be done we find the metals lie on the lower peaked parts and the non- metals on the small flat portions above the line of abscisse. The following results follow :— The metals belong to aggregates having an even number of layers or axes, i.e., the outer rotational motion is opposite to that at the centre. ~) Researches in Vortex Motion. 35 ~ SPIRAL PITCH The non-metals belong to aggregates having an odd number of the same, i.e., the outer rotational motion is in the same direction as that at the centre. In the even series of elements (Series 4, 6, 8) the vortex lines lie between the stream lines and the meridians or, as we may express it, the stream lines lie farthest out. In the metals of the odd series the stream lines lie between the vortex lines and meridians, or the voriex lines are the outermost. In the non-metals of the odd series the vortex lines lie between the stream lines and the meridians, or the stream lines outermost. The metals of high fusibility have their stream and vortex lines nearly co-incident. The alkalis have thetr outer layer thin, the calcium group thicker, and so on. Having fixed their general position, we may now compare with. the curve giving the atomic volumes. When this is done it is found that the atomic volume marches with the moment of angular mo- mentum of the aggregates. In other words— The moment of momentum due to the gyrostatic effect rises and falis with the volume of the atom. All that is yet known respecting the stability of vortex rings leads to the conviction that it is not open to us to explain the various densities of matter as we know it by different densities in the material composing the vortex atoms themselves. We must suppose the matter of all atoms to be the same material as the ether itself. The masses must therefore be proportional to the volumes. It follows that atomic volumes, as ordinarily understood, must depend on the spaces occupied in solid bodies by their atoms. Now a ring will clearly take up more space thau a sphere of the same volume, and we ought to expect high atomic volumes to go with large aperture rings. Combining this with the last result, it would follow that— Moment of momentum rises and falls with the equatorial diameter of the ring atom, which is a highly probable result. 338 Dr. Cash and Prof. Dunstan. Pharmacology of In the present state of the theory, no object is to be gained in pursuing these analogies further. They serve, however, to show directions in which further investigation is to be carried out. It is clear that if a magnetic field is capable of orienting these aggregates, then a substance composed of them will rotate the plane of polarisation of light. “The Pharmacology of Aconitine, Diacetylaconitine, Benza- conine and Aconine considered in Relation to their Chemical Constitution.” By J. THropore Casu, M.D., F.R.S., and WynpHAamM R. Dunstan, M.A.,. F.R.S.. Re- ceived January 13,—Read February 3, 1898. ( Abstract.) The investigation which is described in the present paper has been carried out with pure specimens of the alkaloids aconitine, aconine, and benzaconine, the chemistry of which has been fully studied since 1891, by one of us in conjunction with his assistants and pupils, and forms the subject of numerous papers which have been communicated to the Chemical Society, and printed in the ‘ Journal of the Chemical Society.* As these papers contain a full account of the chemical composition and properties of the various aconite alkaloids, it will not be necessary to do more now than summarise for reference the chief properties of the substances employed in this enquiry. Aconitine is the poisonous alkaloid contained in Aconitum napellus.t Commercial specimens of aconitine vary considerably, many of them being mixtures.{ Until quite recently the pure alkaloid was not an article of commerce. It is a crystalline base, very sparingly soluble in water, but readily dissolved by alcohol. Its alcoholic solution is dextro-rotatory, whilst solutions of its salts are levo-rotatory.§ Even very dilute solutions produce a characteristic tingling and numbness on the tongue and lips. The alkaloid suffers decomposi- tion when heated to its melting point; a molecular proportion of acetic acid is lost, and an alkaloid pyraconitine remains.|| The hydrolysis of the alkaloid occurs in two stages. In the first, which is best effected by heating a salt of aconitine in a closed tube with water,{ a molecular proportion of acetic acid is formed, and an * “Chem. Soc. Journ.,’ 1891—1897. + Dunstan and Ince, ‘Chem. Soc. Journ.,’ 1891, vol. 59, p. 271; Dunstan and Umney, ibid., 1892, vol. 61, p. 385. t Dunstan and Carr, ‘ Chem. Soc. Journ.,’ 1893, vol. 63, p. 491. § Dunstan and Ince, Joe. cit. || Dunstan and Carr, 2did., 1894, vol. 65, p. 176. {| Dunstan and Carr, idid., vol. 65, p. 290. = eS ee a SS oe To Aconitine, &c., in relation to their Chemical Constitution. 339 alkaloid produced which is named benzaconine, the chief constituent of the picraconitine and napelline of previous observers.* Further hydrolysis, by alkalis or acids, resolves benzaconine into aconine and a molecular proportion of benzoic acid, and these are the final products of hydrolysis. A characteristic qualitative reaction of uconitine is the formation of a crystalline purple precipitate of aconitine permanganate when a faintly acidified solution of an aconitine salt is mixed with a solu- tion of potassium permanganate.t+ Most aconitine salts crystallise well from a solution in water, and in experiments on the physiological action of this alkaloid an aqueous solution of the hydrobromide has been employed. Neither the composition nor constitution of aconitine can be re- garded as settled. In determining the exact formula by which the composition is best expressed, there is the difficulty of deciding between several formule which represent the composition of the alkaloid within the limits of experimental error. Alder Wrightf adopted the formula C;;H,,NO,. as best expressing the composition. Later observers, Jirgens,{ Libbe,t and ourselves have so far accepted a formula identical with or differing but slightly from that of Wright, as indicating the composition of aconitine and its derivatives. Recently Freund and Beck§ have proposed for aconitine the formula Cy,HyNO,, instead of that employed by us C33HyNO», since they have obtained from the ultimate analysis of the pure alkaloid nearly 2 per cent. more carbon than was found by Alder Wright and his colleagues, by Jiirgens, by Liibbe, or by our- selves. The question of composition is, therefore, still unsettled and can probably only be finally decided by the analysis of simpler derivatives of aconitine than have been hitherto dealt with. The constitution of aconitine cannot be considered until more is known of the simpler derivatives and decomposition products. For the purposes of the present discussion it may be regarded as acetyl- benzaconine, but nothing is at present known of the constitution of aconine. Diacetyl-aconitine is an alkaloid obtained from aconitine by acting upon it with acetyl chloride,|| and differing from it in containing two acetyl groups in the place of two atoms of hydrogen. It is a erystalline base, very sparingly soluble in water, but readily in alco- * Dunstan and Harrison, idid., 1893, vol. 63, p. 443; Dunstan and Carr, idid., 1893, vol. 63, p. 991 ; Dunstan and Harrison, idid., 1894, vol. 65, p. 174. + Dunstan and Carr, ‘ Pharm. Journ.,’ 1896, vol. 56, p. 122. t Alder Wright and Luff, ‘ Chem. Soc. Journ.,’ 1877, vol. 31, p. 143; Htingens, ‘Inaug. Dissert. Dorpat, 1885 ; Liibbe, Clerk Maxwell’s ‘ Electricity and Magnetism,’ vol. 1, p. 273, edition 1873, it is shown that the deflection of the needle of a quadrant electrometer should vary as (A-B)(c-=S—, where C is the potential of the needle, and A and B the potentials of the two pairs of quadrants. In fig. 1 the H.M.F. between the quadrants was less than 1 volt, and was constant. By the formula the quotient C/@ should in this case be constant where @ is the observed deflection. It varies in arbitrary units from 0°55 to 0-11 as the value of C varies from about 550 to 2,450 volts. This is working the instrument far * See ‘ Philosophical Magazine,’ April, 1885. 308 Mr. E. Wilson. The Kelvin Quadrant hig? a; Lrectrometer Derlectior. Le) 8 is} bolts cor Needle. beyond the range for which it is intended, since when the gauge is in proper adjustment the value of C is only about 550 volts. In the following experiment the highest E.M.F. employed is 115 volts, and since a square root of mean square value equal to 100 volts was the maximum potential difference about to be used by the author in a certain series of experiments upon alternate current Watt-hour meters, it was necessary to see that within this range of potential the formula above given is verified. The instrument was connected as before with one pair of quadrants to the case, the other pair being insulated and the electromotive forces applied to the quadrants, as also to the needle, were supplied by storage cells, and accurately measured by Poggendorff’s method, the standard of com- parison being Clark’s cell. The results are given in Table I. The instrument in the above experiment was mounted on a slate base in the upstairs room of the Siemens Laboratory. The spot of light when working on this base with this instrument is never per- fectly steady, and this may account for the errors observed in Table I. Method of Test. Fig. 2 gives a diagram of connections showing how the electro- meter was used as a Wattmeter for alternate currents, and how it was tested when being so used. In the formula oe MA-B)(C->), Electrometer as a Wattmeter and Voltmeter. 359 Table I. A+B Observed A+B Gosnell WO CER). deflection @. volts. ; eee +106 Gy G5 ince 20°7 436 —140 58 °6 39°5 427 —331 aeue 53 °4 A31 — 552 5 YN “1.6 432 —773 51°5 91:0 A34 —551 32-5 89°7 433 — 268 14.:°2 89 ‘7 438 —352 14°1 115°0 432 —729 32°3 | 114 °0 434 — 726 48 °5 88 °9 | 432 —427 60°4 60°7 | 431 —338 88 ‘9 60°7 4.27 —110 113 O 60°7 432 + 626 SO 32°2 439 + 987 113 °0 18°1 439 433 mean Fig. 2. where \ is a constant, it follows that if A and B are in phase with one another and with the alternate current, and have the same wave form as the alternate current; and if C is in phase with the potential difference between two points of the circuit where the power is to be measured, and has the same wave form, A must be equal and oppo- site in sign to B, since the instantaneous rate at which work is done on or by the circuit must be proportional to AC or BC. _ 7 (ea Va aes 360 Mr. E. Wilson. The Kelvin Quadrant In the Siemens Laboratory there are two alternate currerit machines* coupled together in such manner that any desired phase difference between their armatures can be obtained. In fig. 2, M, and M, represent the armatures of these machines. On the shaft of one of these alternators is fixed a revolving contact maker, M, which makes contact between two brushes once in a period, that is six times in a revolution of the alternator, since there are twelve poles. It consists of a gunmetal disk keyed to the shaft of the alternator, and carrying two rings, one of ebonite and the other of gunmetal insulated from the disk by means of the ebonite ring. Into the ebonite ring are in- serted six contact-making strips of gunmetal one-sixteenth of an inch thick, equally spaced out on the circumference and soldered into the gunmetal ring. An insulated copper brush bears on the gunmetal ring, and an insulated steel brush bears on the surface of the ebonite ring, touching each of the contact-making strips as the contact maker revolves. The epoch at which such contact is made by the small steel brush can be varied and observed by means of a pointer moving over a fixed circle divided into 360 equal parts. The diameter of this revolving contact maker is 13 inches. Q: is the No. 184 electrometer used as a Wattmeter. Q, is the No. 71 electrometer used in connection with the revolving contact maker M for the purpose of determining the instantaneous values of the current and potential difference. D is a Kelvin balance or Siemens electro-dynamometer for the measurement of current; H is the thick wire circuit or circuits of the Watt-hour meters being tested; F is a Kelvin multicellular voltmeter ; 71,7 are non-inductive resistances of comparatively large value for the purpose of reducing the potential difference applied to the electrometer Q2, when measuring potential difference C; the pressure circuits, P, of the Watt-hour meters are placed across Titre; 73,7, are made up of a manganin strip 50°8 mm. wide and 0°4 mm. thick; 73 = 74 = 0°2275 ohm at about 10° C.; 7;, 75 are non-inductive resistances of considerable magnitude for reducing the potential difference applied to Q. when necessary. The junction between 7; and 74 is connected to the case of Q,; the quadrants of this instrument are connected respectively to the extreme ends of 73, 74; Whilst the needle of the electrometer is connected to the other pole of M;. Im connection with Q., 8: is a two-way switch for observing potentials across 72 or 7;; S, is the ordinary switch sup- plied with the electrometer which short-circuits the quadrants when moved to its central position, and in its two other positions revérses the charge on the quadrants; Gis a condenser, which can be varied * A full description of these machines is given in the ‘ Phil. Trans.,’ A, vol. 187 (1896), p. 281. ” +S Electrometer as a Wattmeter and Voltmeter. 361 from 0-001 to 1 microfarad, its capacity being 1 microfarad during the experiments, the results of which are given in Table II. Before giving the results of the experiments it is well to explain the method adopted of treating the curves for the purpose of arriving at the average Watts due to the alternate current, the relation be- tween which and the deflection of the electrometer used as a Watt- meter it is desired to find. It is also necessary to examine the limits of accuracy obtainable by this method. In any one experiment the frequency employed is kept constant as nearly as possible: the phase difference between current and potential is adjusted to any desired value and the amplitude of these quantities is kept constant by ob- “serving their square root of mean square values on the instruments Dand F. The revolving contact maker M is then set to different positions of the phase, the number employed being at least ten equal divisions to the half period, and for each position, readings taken on the electrometer Q, when the switch §, is in each of its two positions. If the deflections so obtained be plotted in terms of the position of the revolving contact maker M, the forms of the two curves are those due to the instantaneous values of the potential difference applied to the needle of the electrometer Q,, and the current which gives the form of potential difference applied to the quadrants of Qu. By multiplying each of these deflections together, and by a suitable constant involving the square of the sensibility of Q, and the resist- ANCES 7), 723 73, OF 73; 15, 7s, the instantaneous rate at which work is being done by the alternate current can be inferred in Watts. The average of these over a half period gives the average rate, and this ean be obtained by plotting the instantaneous product and taking the area with a planimeter, or the average of the algebraic sum during a half period can be taken. The author found the latter method agreed so well with the former when the number of inter- vals at which observations are taken is ten, that he has adopted it in this paper, that is to say, the two electrometer deflections for a given position of M are multiplied together, the average of these taken over half a period, and such average multiplied by a constant to reduce to Waits. The best way, perhaps, to test the limits of accuracy is to adjust current and potential until they are exactly in phase. The volt- meter F and amperemeter D give the square root of mean square values, and the product of these should agree with the average results obtained from the curves. The time required to take one set of observations is generally about twenty minutes, during this time an average for volts, amperes, and frequency is taken. The author finds from experience that if care be taken an agreement between the results got from the curves, and from the product of volts and amperes, can be obtained to within one or two per cent. It 362 Mr. E. Wilson. The Kelvin Quadrant must be remembered that for each position of the contact maker, four observations on the electrometer (Q2) scale have to be obtained ; that is, two for potential and two for current corresponding to the two positions of 8, for each position of S,, the difference in each case giving the net double deflection. This method is best, as it eliminates any zero error there may be. In working the electrometer Q., a wooden tapper or mallet is employed, since in every electro- meter there must be viscosity due to the fluid, and by gently tapping the slate base for each deflection very consistent results can be ob- tained. This viscosity is greater in winter, and it is advisable to keep the instrument in a warm room, although with this method of tapping the author does not find this necessary. The greatest trouble in the use of the electrometer undoubtedly arises from dust settling on the surface of the acid in the jar, thereby making the angular movement of the wire hanging from the needle smaller than it would be if such brake action did not exist. This takes place when the acid in the jar is old, and if the surface be agitated by blowing through a glass tube near where the wire dips into the acid it can be to a great extent remedied. Whatever the state of the acid the author finds he gets the most consistent results by gentle tapping. The electrometer Q, is not so sensitive as the old form Q),, and the effect due to the acid in it has not given so much trouble. The sensibility of Q. when the idiostatic gauge is adjusted is such that one Clark cell gives a deflection from zero of 105 inches cn a scale 12 feet from the mirror. The potential of the needle is in this case about 350 volts. | Experimental Results. In making a thorough test of the electrometer as an alternate current Wattmeter we have the following variables to deal with :— 1. The frequency of the alternate current. 2. The phase difference between current and potential, that is between C and A or B. 3. The amplitude of C and A or B. 4, The shape or wave form of the curve of potential and current. The results obtained are tabulated in Table II and are divided into three groups (a) (6) (c). In group (a) two frequencies are given, namely 41°6 and 83 complete periods per second. The potential on the needle is constant at about 100 volts (,/mean’). The phase difference between potential and current and the current itself are each varied. When the phase difference is zero, it is only necessary to take the product of the ,/mean* values to deduce the Watts, although in section (b) three instances are given in which for phase difference zero, the Watts are deduced by both methods. The Llectrometer as a Wattmeter and Voltmeter. 363 average Watts per division given. by section (a) are 17:00 for all angles of phase leaving out the two values deduced by aid of the cosine law for angles of 30° and 60°. It will be seen that under the con- ditions of section (a) the Wattmeter may be said to be verified within the limits of accuracy attainable by the method of test. The wave form of the unloaded alternator is given in fig. 3 and marked C; this is the wave form of potential applied to the necdle in all experi- ments in sections (a) and (b). A sine curve having the same maxi- mum ordinate is superposed for the purpose of comparison. The current curve has different wave form according to the load on the alternator. For small currents it approximates to C in fig. 8. The eurve A, fig. 3, is the wave form for current 74 amperes, which is the maximum we have employed. Bre.3. TS SRR Ree ee ppp Eee sear daas/4a558\eS561 56 | 1 GS RB aa ee RRR i fHHH-- i Bag The experiments in section (h), Table IT, are intended to demon- strate the reliability of the instrument when the potential of the needle C is varied through wide limits. One would expect from the curve in fig. 1, that for high potentials on the needle the Watts, per division of the scale, would diminish. This is found to be the case when the potential C is raised to 1,860 volts (./mean?) for fre- VOL. LXII. 2-0 oe" >see The Kelvin Quadrant Mr, &. VWuson. & ’ at ‘MBIT OUISOD Aq pooNpep §}3UM x F daqmesag | LL-ST L9.ST 0. 9849 0. 6899 68: II 0- 69S ST ft) BB (2 Pp dtequieoeq | 62- LT uf 0- L60F a | GI. FL 4-101 0-09 0- &8 Me TeqmrekON | -2OV kT te 0- 2291 Ng 6-68 0- OOT 6-89 0- &8 : - re LL ng 0- 1618 ag GL-OF P 66 &- 6E 0-€8 T 2 g. 41 aS €- 368 0. 896T 94-61 9. 66 6-19 0- €8 : 2 aS 89. 9T “ 0-ZLPL 86- 84 O- IOT 0-0 0- 88 4 a G8. 9L *s 0. S268 GP.6E GS. 66 0-0 0-8 G : ee L@- LT es GS. P86 OL-6 g. TOT 0-0 0- €8 : rt 06- 9T a 9- 086 F9-6 | 4-101 0-0 0- &8 TESTES T01 : ¢8. OT ye | 0. 166% IG xceuees| 0- OOT 0.0 O0- &8 0G @ 48-91 ee | Q- 4688 FG: 68 €- 66 0-0 0. €8 : e #PS- ST “ | ue L- 68 0- 00T P.6S - 9. 1F come e #86. LT "s 2 9. 68 0- OOL 0-08 9. 1F : vs 8. LT me O- LSbT on C8. OT 99. 66 G.18 G. 17 &Z Gg. OT Bs O-1L88 | 0-966 G8 -68 €- OOT 1. 6% 9. IP cee e 64-91 oh | 0: 9668 8- 68 | ¥-OOT 0: 0 9. IF e ” GL. 9T te On60LE IL- AL £8. 66 0 9.1F ¥S L9G wWaAao Ny a 64. 91 * | 0-6246¢ G9 .68 Gg. OOT 0-0 a i D *sdAano eee te *SdAINO een | “L6O8T AqMO ST) qonpord wLo01 7 pai uents) qonpoid mor | *,ureut yh | ‘s]]OA UBT A ‘potod T == .09¢ Staal ‘quoutted xa ae 3 “sodedme Ul O a[peou ‘seaIsop UL “aL yo. SLES UI qUaTING jo [etwusjz0g QoUTO HIP OSBY "e[Bos rey | Jo worstarp aad 5370 \\ me | | ‘TI 198, Electrometer as a Wattmeter and Voltmeter. = - = Z8- OT - 0- 1F2z # 0. SF g- ZIT CT toquiavad | GI. LT T6: ST 0- OES 0. GLES 0- SF 6: LIT é 66 ee ee | : | | 09- TL O- 0686 0€-T ' Q- OF8T LT 4 12-11 Fes 0- 0OFF 68: | 0: OF8T # 6F- OT oe 0. 8982 al 10- ST 0- 98F 6 = 68. 9T ZS: OT 0- Z9TZ 0. 0202 £0- OT | 0. S&P : oa OV- IT ae 0. OGES Gc: 1 | O- O98T as Bi €6- OL 2 0: O&eF ge.Z | 0: O98T ve yh Oe a 0-&&hz ns &- O01 0- 829 TI = L&- OT GC. OT O- Z689 O- GE89 98: OT 0. 0€9 py doqurs0a(T Cs, any hs will only (neglecting terms of order e”) be correlated with its own particular (7 —s)th parents, male and female, and there will be $(2”-*) such male parents and $(2”-*) such female parents. Hence N 3 : Nx ZsZupns = ostn {oso 2irt Si(7sn) +16 56,25" So(7'sn) ’ = pay + MOso 2° 18 3( ton) a= Na sa 2° tS ays Pus — Need Sur sn) +m = ra 8, (7'sn) $m S3 (7 sn) + — on erly sn) } i Reet esse an ae ee Ses a wee e-ecuret a) CUD) Here §,(7sn) is the sum of all 75, which begin and end with male in the descent, S2(7s,), of those which begin with female and end with male, $3(7sn), of those which begin with a male and end with a female, and S4(7sn) of those which begin and end with afemale. Now, as before, put m = o;/o's = on/o',. We thus have, supposing the varia- bility of each generation to be constant, Psn = it + és gr stl per ae = a OMe Matha cate kee USS a Gv), 1+ és; where 7;, now stands for the mean value of all the correlation co- efficients of an individual and its individual (n—s)th parents. It may be written r,s, as it depends only on the dzfference of the gene- rations. Hence supposing sexual selection to remain constant, if it exists, for all generations, we see that p,; depends only on the differ- ence of generations, and may be written py_s, or: Pu—s = Oeil (1 oF €) ae Now if there be no selective breeding, e appears, at any rate for man, to be small. Hence we have the important proposition : The correlation between two mid-parents, p generations apart, is equal to the product of 22? and the mean of the coefficients of correlation between an individual and its individual pth parents, when they are taken for all possible combinations of sex. When no allowance is made for reproductive selection, it has been shown by Miss Alice Lee and myself that the four possible 7’s * The importance of this result is that it reduces the nue 1) correlation coeffi. cients between ” mid-parents of different orders to coefficients only. 390 Prof. Karl Pearson. for first parent and offspring are very nearly equal;* assuming the equality of all possible 7;’s for the sth parent and offspring, and neglecting e we have or, we conclude that very approximately: The standard deviation of the mid-sth parent may be obtained from the standard deviation of indi- vidual sth parents by dividing by 235, and the correlation between mid- sth parents and mid (s + p)th parents may be obtained by multiplying the correlation between an individual and any sth parent by 2’?, Thus the variability of the ‘sth mid-parent rapidly decreases as we increase s, 7.e., as we get back in ancestry the mid-parent comes more: and more nearly to represent in all cases the mean of the general population. Whether the correlation tends to decrease or increase will depend on the relative rates of change of 22” and 7p. Since pp must always be less than 1, we obtain at once the interest- ing limit that the correlation of an individual and a pth parent is. always less than (0°5)#?. For example the correlation between : Offspring and parent must be lessthan 0:71 i, and grandparents Be a 0°5 is and great-grandparents 3 a 0°36 i and great-great-grandparents _,, fs 0°25 Their actual values as deduced from Mr. Galton’s law are much smaller, as we shall see later. (3) The reader will remark that in order to get these results in a simple form we have multiplied the female deviations from the mean by a constant factor m, which has afterwards been taken equal to the ratio of male to female variability. The reason for this was two- fold. In the first place o is certainly not equal to a’, and, conse- quently, m= 1 would not have given il me 2 Ce, OLE —= aan Vozto? f 92(s+1) =, = 36 a more complex form. In the next place we note the fairly close equality of 7’, 7", 7'", 7", when we neglect reproductive selection ; hence m= a;/o'; is the only value which appreciably reduces formula (iii) as well as formula (i). I therefore define a mid-parent to be one in which the deviations of the females are reduced to the male standard by first multiplying them by the ratio of male to * © Roy. Soc. Proce.,’ vol. 60, p. 278, Mathematical Contributions to the Theory of Evolution. 391 female variability. This does not theoretically agree with Mr. Galton’s definition, for he reduces the female to the male standard by multi- plying them by the sexual ratio, or the ratio of the male to the female mean for the organ under consideration. In order, therefore, that my factor of reduction should agree with Mr. Galion’s, it is needful that the ratio of the standard deviations should be equal to the ratio of the means, or, in other words, that the coefficient of variation should be the same for the two sexes. Now for the stature of men and women, I find for 1000 cases of each sex the coefficients 4-07 and 4-03 respectively, or the coefficient of variation is sensibly equal for both sexes.* Mr. Galton found from his anthropometric laboratory returns for somewhat fewer numbers, and probably for a lower social class, values of 3°75 and 3°79, again sensibly equal. Hence the mid-parent, whether defined in my manner or in Mr. Galton’s, would have a sensibly equal value in the case of stature, which is the one Mr. Galton dealt with in his ‘ Natural Inheritance.’ The coefficient of variation is, however, not the same for both sexes in the case of all organs,t hence for the purpose of simplifying the . formule, I am inclined to think my modification of Mr. Galton’s original definition will prove of service. (4) I shall now proceed to determine by the law of ancestral heredity the correlation between an individual and any sth parent from a knowledge of the regression between the individual and his mid-sth parent. By the principles of multiple-correlation if a, a, a, ....#n be m+1 organs, with standard deviations op, o, o.....¢ , and correlations Tory T02) T03 0 e #6 112) T13 «+ +* Tn_1, ny then the frequency surface is given by 1 { (2 y 2\2 (= zy) \ ——<~R 0) +R 1) +... ee $2Ro( OL +. te z= constX e 2k 7 a) : (>) : 0192 where R a d i To15 To25 To3- eaoeoe Ton ~ Tols 1, T3125 1139 e+ee Tin Tony Tins Ton cevsesesce Tan { and Ry, is the minor of the constituent of pth row and gth column. * See ‘The Chances of Death,’ vol. 1, “‘ Variation in Man and Woman,” p. 294. + ibid., p. 311. Mr. Galton’s family record data gave 1°032 and 1°005 for the ratio of the coefficient of variation of sons to daughters and of fathers to mothers respectively. See ‘Phil. Trans.,’ A, vol. 187, p. 271. ~ Many cases are given in the paper on ‘‘ Variation in Man and Woman,” cited above. 392 Prof. Karl Pearson. Putting x, %,..-.#n constants, say hy, ko, .... kn, we have for the mean value % of the corresponding array of 2’s, as ier om) Roeeo Rose 0 Rone 7 k zi Weg te Ghee Tey The standard deviation of x for this array is wa Eee eoreecereeee reo eeoee (vil). These results (vi) and (vii) are the regression formule.* Now let 2, %...., be the mid-parental values of the lst, 2nd, ord, .... uth order, and % = hk) the mean value of the organ in the Hee oe Then the value of R is given by R — | Ls Pls P29 P39 Ph eece Pn Pry 1, Pry Px» P38 ee ++ Pai Puy) Prat “e's ss 5a ee onan 1 and the regression formula is: ives (5. oa Ape Go Ron Se pea ks. eee kn 9 hs, i Roo Ze Roo = if we stop at the nth mid-parent. Comparing this result with the analytical statement of Mr. Galton’s law of ancestral heredity given on p. 388, we see that we must have from (v): i> i ae Re Re ee eee crf Boo oa 2/2 | Roe/Roo re = -(5>5| 4 Ee can avons (ix) There will be 2 such equations, if we go to the mid-nth parent, and there are » quantities pi, p,....px to find. Thus Mr. Galton’s statement that the partial regression coefficients are 4, 7, }.... * See ‘ Phil. Trans.,’ A, vol. 187, p. 302. Mathematical Contributions to the Theory of Evolution. 393 gives us sufficient equations to find the coefficients of total correlation between the offspring and the successive mid-parents. Hquations (v) will then enable us to find the coefficients of correlation between an individual and any individual ancestor. But these in their turn will suffice to determine all inheritance whether direct or collateral (see below). In short if Mr. Galton’s law can be firmly established, 7 zs a complete solution, at any rate to a first approximation, of the whole problem of heredity. It throws back the question of inheritance upon two constants, which can be once and for all determined; herein lies its fundamental importance. I must confess that this element of simplicity was at first my chief difficulty in accepting the law as laid down in the paper on Basset hounds, and I even yet have a certain hesitation, owing to an apparent difference in collateral heredity in different social classes, and also to the apparent numerical value of the inheritance of fertility in man. (5) I shall next obtain a solution of equations (ix). Taking the minors of the n+1 constituents of the first row, namely, Ro, Ro, Ry, -. -- Rox, and multiplying them in succession by the constituents of the 2nd, 3rd.... »+1th row of R, we have by the ordinary theory of determinants the system : PiRoo = Ro + piRos aie P2kog Bieta esters + pnRon = 0. prRoo che pio =F Ro =F pikos aS SSA eeSs Pn—2Ron = 0. Pqkoo a Pg—1 Ra a Pq—2lRoz + pg_sRo3 + coco t Pu—qhon = 0. PuBRoo as Pn—1Ro1 3 Pn—2 Roe + Pnu—sEog Sar ee Ron = 0. Divide each of these equations by Ry and let us use instead of (ix) the somewhat more general system which will allow us to consider one or two limiting cases, and rather more generally than Mr. Galton has done “to tax the bequests of each generation,’ as he. expresses it :* R : : R., = —1P; ——= —7iP, =: —7P, de. .. (x), 00 0 -where y and f are two constants; we then find: * © Natural Inheritance,’ p. 185. 394 Prof. Karl Pearson. —ptyB+yPatyPpt ..06 HYB"pn1 = 0. —patyPatyP+yBpit 2.66 FYB"pn—-2 = 0. —patyBpg-1+ YE pg—-2 + VP pgs t ++ +e + YB" Png = O. —Pqt1 = YB Pq in 2 pg—-1 air 8? pq—2 = Sie siege YB" pPu—q—1 = 0. — pat BpnratB'pn2+ 1B pns+ +++ +B" = 0. Multiply the q+1th equation by 1/f, and subtract from the qth ; we have 1 B Poti pg +4) oe YB" pu—d == 0 2 90 ese ieee @e (aye Assume py = cz’, hence: a Rae Basi +y¥) + ya 4 = 0. But since @ and f are both less than unity, the last term will be vanishingly small when 7 is indefinitely large, thus: a= PO+9) «cate eee Be et XA) Substituting py = cx! in the first of the equations for the p’s above, we have: | Or, taking as before (#8)” = 0 for n very large: = Mindi ae? 1 a Ciice ( Bele bs 5) Sel ea Ca = ag (12a) Sete CRORCROLONG OTco ° (x11) (P % (xii) and (xiii) contain a complete solution of the fundamental equations for the p’s given above, so long as we go only to a finite number of mid-parents, 7.e., gq may be very large, but not comparable with » = &. (6) Special Cases. - (a) Puty=1, 8=4. It follows thata=l,ande=1. Henceif Bog ot 9 Roo at Mathematical Contributions to the Theory of Evolution. 395: all the total mid-parental correlations would be perfect, and, there- fore, any one mid-parent would suffice to fully determine any other and the offspring. The individual parental correlations would then be 1 1 1 1 We OQ defer 4 for parent, grandparent, great-grandparent, &c., with offspring.* (6) More generally, suppose any values of y and 8 which lead to ¢ = )>then “1-8 0429) whence we find #(y+1) =1, that is, a=1; or again, all mid- parental correlations are perfect. Thus, as in case (1), the individual parental correlations could be represented by These are the values I took in my memoir of 1895. I took these values then because they seemed to express Mr. Galton’s method of passing from individual parental to individual grand-parental total regression.t JI had not perceived that there was any antinomy between Mr. Galton’s theory of regression and his law of ancestral heredity. Had I done so I should certainly, at that date, have given the preference to the former, and rejected his law of partial coeffi- cients of regression in favour of the values, based on numerical observation, of his total regression coefficients. (ey Puthy = 1, B= Aa this is Mr. Galton’s form of the law. Df 2 We find at once a vin Zz 0°6 = = C= =] = Vb, V2 5 Hence we have for the successive mid-parental correlations p:, p2, £39 &e., Moai e |O8 2, ye and for the individual mean parental correlations, 1, 72, 73, &. &e. (35. 015. 0°075,.. we. * This is what, I think, must follow from any theory of the “ continuity of the germ plasma,” and of its exact quantitative addition and bisection on sexual repro- duction. + ‘ Phil. Trans.,’ A, vol. 187, pp. 303-5: ~ See ‘Natural Inheritance,’ p. 133, Mr. Galton puts r= +4 for a parent, * = i for a grandparent, and so on. Y 396 Prof. Karl Pearson. Here med tr tr "4+r'"), %,=158@), Ge Further, for the regressions on the mid-parents (not partial but total), or pr, pr, ps—, &e., we have, on the assumption that all 2 22 2s generations are equally variable, 06, 06, O06, &e. Or we may express the law of ancestral heredity in Mr. Galton’s {orm in the following simple statement :—The total regression of the progeny on the mid-parent of any generation ts constant and equal to 0°6. _ Let us see how these results agree with observations. - Mr. Galton* tells us -that his first estimate of mid-parental regression was 3/5 = 0°6. This estimate exactly agrees with theory. He after- wardst changed the value to 2/3 = 0°67, which is less in agreement. My own calculations,t on Mr. Galton’s data, give 7’ = 0°3959, ry! = 0°3608, 7," = 0°2841, 7," = 0°3018, or 7, = 0°3355 instead of 0:3. The probable error is, however, 0°026. If we do not weight fertility the parental correlation§ = 0-41 + 0°03, a value which is distinctly too high for Galton’s law. It must be remembered, how- ever, that our deductions from that law are based on equality of variation in each generation, and that this equality is by no means the fact. I hope shortly to get final values for parental heredity from my family measurements, which have now reached a total of nearly 1,100 families, and thus settle how far Galton’s law needs to be modified. On the whole the confirmation obtained from stature data for the law of ancestral heredity is very striking;|| I am inclined to think even more convincing than that obtainable from the Basset hounds, and this for a reason to be considered later. It suffices here to observe that we donot need to know the characters of parents, erand-parents, great grand-parents to test Mr. Galton’s law; any single relationship, near or far, direct or collateral (see below), will bring its quota of evidence for or against the law. It will be seen that the table (p. 397) differs in principle from Mr. Galton’s on p. 133 of his ‘ Natural Inheritance.’ In particular, supposing equal variability for all generations, the individual grand- parental regression is not the square of the parental regression, but the half of it. Mr. Galton’s law of ancestral heredity contradicts * ‘Natural Inheritance,’ p. 97. + Lbid., p. 97. t ‘Phil. Trans.,’ A, vol. 187, p. 270. § ‘Roy. Soc. Proc.,’ vol. 60, p. 279. || Good evidence in its favour is also to be deduced from the inheritance of the cephalic index. See paper by Fawcett and Pearson, infra, p. 413. Mathematical Contributions to the Theory of Evolution. 397 Table of Heredity according to Galton’s Law. Individual parent. Mid-parent. Order Correlation and regression. Correlation. Regression. 1 0°3000 0 °4243 0°6 2 0°1500 0°3000 0°6 3 0 :0750 0°2121 0°6 + 0 °0375 0°1500 0°6 5 0°01875 0°1061 0-6 6 0 009375 0°0750 0°6 c@eevesn e ee eevee ee 1 ee eee ae eeoe eevee He gth 0-6(5) O66) el Remarks.—The correlation of the individual first parent is to be taken as the mean of the four possible parental correlations due to differences of sex, if these are not sensibly equal, and a like rule holds for the individual sth parent. The individual parental regression is based on the assumption that the variability of offspring and parent are the same. In dealing with the mid-parent, female deviations must first be reduced to male by multiplying them by the ratio of male to female variability. his views on regression, and it is the latter which, judging from both theory and observation, I now hold must be discarded.* (7) Mr. Galton’s law gives us the partial regression coefficients when all the mid-parents are known. It is desirable to deduce from the theory of multiple correlation the values of the partial regression coefficients when we take 1, 2, 3, 4,....mid-parents only. When g mid-parents are taken let ie partial regression coefficients be e14, ex, Esq, €49, ++ +» €qq; then again we have for the mean of the offspring h,: Oo Oo Go . i — Cig aon in a £24 Te ko+ eeoeve oa €qq_ ky oeeeesn (xiv), oO 02 oy where ois the standard deviation of the offspring and o, of the pth parental generation. Comparing this with the regression formula immediately under (viii) we have Ro -z B= — 2b OY, by (v), Rez i * T do not agree with the last column of Mr. Galton’s table giving the variability of arrays. For single correlation the variability (standard deviation) of an array = ¢~/1—7", where r is the correlation and not the regression. With equal varia- bility of all generations, rin the case of the individual parent may be replaced by the regression. But the correlation is not equal to the regression in the case of mid-parents, because the variability of the mid-parent by (v) is increasingly less than that of the offspring. 398 Prof. Karl Pearson. Now make use of the general equations for the p’s given just below equation (x), substituting for the R’s in terms of the e’s, and remem- bering that we are to stop at n = q, that py = cz?, and that 1/,/2 =<. ‘We have after some reductions the system : i z Wk 4 6 2qg—2 pies 1 2 4 Yg—4 i= ara Eg ta €3g + & egg + eee TO Egqy U 1 — By tee t €3g +a? eqg+ eeee +a eqg, Cr ee} 1 ‘— €1g + €ag + €3g + cece +e eq-1 gta egg, : if 1 = e\ygteagtesgt---- tT eg—1 gt Egg Subtracting the (q—1)th of these equations from the qth, the (q—2)th from the (q¢—1)th, &c., and introducing the values of a = +t and of c = 0°6, we find O°4 €g_1 g—O'7 €gg = 0. O°4 egg —0°7 €gas g—- O'S Egg = 0 . nee oe oe a (xv). O°’ €7-3 g—O'7 Egg g— 0°15 € g_1 g —0°075 egy = 0. O'4 €y_4 g—O°7 Egg g— 0°15 € g_2 g—0°075 €g_) g—0°875 Egg = 0. ‘and so on, each new coefficient being now half the last. These equa- tions give successively the ratios of e7_1 g, €g-2 q ,€¢-3 q, &C., to egg. Hence the last of the previous set of equations will then give egg. Thus the partial regression coefficients for any limited number of mid-parents can be found. This last equation also gives us ibe S(e) —a Eqg = iia Eggs a convenient formula for measuring how nearly the mean offspring of g mid-parents, all selected with a peculiar character, k; = k, = ks =.... = hk, = K has attained that character, For in this case ity = S(ek) = K x S(e), and oo] I= 1-= €gg «eos 4a (xvi). - Mathematical Contributions to the Theory of Evolution. 399 Hence the more nearly 1—«,, = unity the more nearly the offspring has the full character of its selected parentage. I venture to call this expression the stability of the stock. It is a measure of the stock breeding true. Lastly, to find the standard deviation of the array = o/R/Ro we have only to express R in terms of the minors of its first row, or, R = Rot piri t peRoe -. +». +pgRog: R/Roo = 1— C (eyga” + Cog! == Egg” + oe ee + Egg!) ss : ; 1 ee = 1—0°3 (agts ee eee. ee €qq) pao acc (xvi). In the limit when g = co R/Ro = 1—0'3 X40 +44+44+.----) =1—02= 08, and / R/ Roo = 0°8944. The following table has been calculated from these formule :— Table of Pedigree Stock according to Galton’s Law.* Ratio of Partial regression coefficients, Be variability of _ | Stability. offspring to Sions | tat of whole S («). population. &}. Ep. E3. Ey E5. EG. 1 0 9055 0°6 — — — | _~ == 0 °6000 (05) 2 0 °8946 0 °5122 | 0°2927 —— == a Bee 0 -8049 (0-75) 3 0 °8945 0°5015 | 0°2553 | 0°1459 _ — — 0° 9027 (0875) 4 {| 0°89445 0 °5002 | 0 -2507 | 0°1276 | 00729 — a O:9514 | 3 (09375) | 5 0 °894.4, 0 *5000 | 0:2501 | 0°1258 | 0°0688 | 0:°0365 — 0:9717 (0 9687 6 0° 8944, 0 ‘5000 | 0 :2500 | 0°1250 | 0°0627) 0°0319 0 ‘0182 09879 (0 9844) ee — ——_ — er — —S —_— 00 0 °8944: 0 °5000 | 0 °2500 | 0°1250 | 0°0625| 0:°08128 | 0°015625 1 * To save possible labour, in case it should ever be needed to investigate the partial regression coefficients for more generations, I place here the ratios of the first six e's; : €q-1 q/€aq = wir Eq—2 gl€qq = 3°4375 ; €q-3 ql€aq = 6°859,375. €q—5 q|€qq => 13°714,843 ; €d—6 q/€aq = 27°428,709. 400 Prof. Karl Pearson. (8) I venture to think this table of considerable suggestiveness, and will now point out some of the conclusions that may be drawn from it. (i) With a view of reducing the absolute variability of a species it is idle to select beyond the grandparents, and hardly profitable to select beyond parents. The ratio of the variability of pedigree stock to the general population decreases 10 per cent. on the selection of parents, and only 11 per cent. on the additional selection of grand- parents. Beyond this no sensible change is made. - We cannot then reduce variability beyond 11 per cent. by the creation of a pedigree stock, z.e., by breeding from selected parents for 2, 3, 4,....” genera- tions. In some cases of course we appear to decrease variability— for example, if we increase the average size of an organ—for the absolute variability is then asmaller proportion of the actual size, and the relative variability, or coefficient of variation, may thus be steadily decreased. If Mr. Galton’s law be true, then pedigree stock would retain only a slightly diminished capacity for variation about the new type. For example, the absolute variability of men of average height, 69°2 inches, being 2°6 inches, the absolute variability of men of 72 inches, obtained by selecting any number of 6-foot ancestors, would hardly fall short of 2°3 inches.* (ii) Two different classes of pedigree stock exist. In the one we start with the general population, and select special characters for 1, 2, 3,....% generations. In the other we know the pedigree for 1, 2, 3,....” generations, but have no reason for supposing that before these generations the stock was absolutely identical with the general population. In the former case we put for the mid-parents by — Whe = Wz == 2 eos =e King = hint, = « «(et ee Os Hence the regression formula is if i 1 la=( ERE... +3) as eR The values of k,/K are tabulated in the last column of the table above in brackets. They give the ratio of character in offspring to character in ancestors, if ancestors of equal full character have been selected for 1 generations. We see that in six generations the off- spring will have been raised to within 1°6 per cent. of the selected ancestral character. In the latter case we must use the partial regression coefficients. * The probability of an individual of selected stock differing widely from the pas is of course much less than in the general population, because the stock is, as a rule, far less numerous. Mathematical Contributions to the Theory of Evolution. 401 €}, €.... of the table. For example, in the case of Mr. Galton’s Basset hounds, 0°5015, 0°2553, and 0°1459 were the coefficients to be used, rather than 0°5, 0:25, and 0°125, when he proceeded to apply the law to three generations. These give the proper allowance for the ancestry beyond the pedigree. Thus the great-grandparents ought to have been given about a fifth more weight. If we proceed to six generations in pedigree stock of the latter type then the offspring will be within 1:2 per cent. of the selected ancestry, 1.e., their stability as given by the last column = 0°9879. (ii) Now let us apply these results to the all-important problem of panmixia and degeneration. Suppose a selection made of a par- ticular character for n generations, starting from the general popula- tion. Then the offspring in the (n+1)th generation will have 1 — of the character on the average. Now, stopping selection, Jet us breed with a first generation of mid-parents with Bi Eo the character. on The offspring will have: i it = (3 —i)+4 —" AET: oe tees = (1-3 )+ +3(1- : ay == of the character Ps Qn ani} The n+ 2th generation will have: Mie tt Y> I trash nd 1 - oe —( 1—— J4-+—4+....+— 3 ( QF 197 ( z)+Etat 1 53 Z 1 1 1 z 1 au! = (ima) ts(-a)ta (ta) ee on of the character, and soon. The law is obvious; the offspring will always have the same amount of the character as had the generation after selection ceased. If we start with pedigree stock with wnknown ancestry beyond the nth generation, we reach the same conclusion. Thus, after three generations the offspring will have 0:9027 of the selected parents’ character. Now stop selection and the fourth generation will have: 0°9027 ey et este = 0°4515 + 0°2507 + 0°1276 + 0°0729 = 0°9027, the fifth generation will have 0°9027 (e+ €2) + est e+e5 = 09027, again, and soon. The general law is obvious. VOL. LXII. 26 402 Prof. Karl Peargon. Thus, on the basis of the law of ancestral heredity the case against panmixia is even stronger than it appeared in my memoir on heredity.* Assuming Mr. Galton’s law of regression, I there showed that panmixia was possible with a stable focus of regression, but that the supporters of the consistent theory of panmixia must place that focus of regression, in order that degeneration should be continuous, in a position inconsistent with observed facts (p. 314). We now see that with the law of ancestral heredity even this is not possible, a race with six generations of selection will breed within 1:2 per cent. of truth ever afterwards, unless the focus of regression instead of being steady actually regredes. Of course there are many ways in which this law may be modified. For example, fertility may be a maximum with the average, say, of the unselected original population, and after a selection it may remain correlated, having the lesser values of the selected character more fertile than others.t Then, of course, the stock would degenerate with panmixia.t This would, however, be reproductive selection, not panmixia in the ordinary significance, reversing natural selection. We are far too ignorant at present of the correlation of fertility with other characters to base any sweeping principle like that of degeneration by panmixia upon it. Our attitude at present can only be that there are no facts, and that there is no workable theory of heredity yet discovered which favours in any way degeneration by panmixia. (9) Taxation of Inheritance—If we assume Mr. Galton’s law of ancestral heredity to be a limiting statement, we can at once from our general formule ascertain the influence of “taxing the irherit- ance” in any other than Mr. Galton’s form. He has, in fact, taxed the inheritance (where by “inheritance” I understand deviation from the mean of the general population, not actual size of the character), 50 per cent. in each transmission. ‘There may, however, be two types of taxation, a general taxation on the individual receipts and a special tax on each transmission——corresponding, so to speak, to a duty paid by an individual on coming into receipt of the entire ancestral property, and a stamp duty on each conveyance of an individual ancestor’s contribution. The first 1s represented by the y of our equation (x), and the second by the /2,. Mr. Galton, in his memoir on Basset hounds, has stated certain conditions of the law of ancestral heredity, and he concludes (p. 403) that his conditions are only fulfilled by the series gt (a) ay coiee ee * ©Phil. Trans.,’ A, vol. 187, p. 308 e¢ seq. + This is how I should at present account for the degeneration of pedigree wheat. + Some influence of this kind is possibly sensible in highly civilised communi- ties. See ‘Reproductive Selection,” in my ‘Chances of Death,’ vol. 1, pp. 98 e¢ seq. ~ ‘ Mathematical Contributions to the Theory of Evolution. 403 It seems to me that they are equally well fulfilled by the series qB +yB?+9B°+ eee es provided the sum of this series is equal to unity, or y8'/(1—yB') = 1, that is 1p’ = 4. Pe im Rog Sp NY But yok = Say 70 > = yt X (4/2) by Ge); 00 ki. +B" = het cose 1 * The appearance of the 2 here requires notice. Let there be v brothers to the array of any single series of mid-parents; then if s be the standard-deviation of the array, the distribution of brothers corresponding to a given ky (ea a(e!/ 8). Vv / Ins Hence the frequency of a brother between x’ and x’ + dx’ occurring with a brother between x” and x2” + dx’ 2 Vv — 1(y!']5)2 —1(!1/9)2 Se edge eB 1 Gal’, V 2s? If we take as limits 2 and x2”, both = + to —, we shall clearly take each drother twice over with each other brother. Hence— 2 LS (al /e\2 ! , 5) S(x, x2) = 4 meh = : (eo + 2!)(Keg + a’"e7 2 © /3°+(a"/sp 5 dx’ dx!” =3v7k,?. TS —oJ—x Now allow one pair of brothers to each system of mid-parents, and S (a2) =2h,? for one mid-parental system, or if there be z such mid-parental systems, S (2.2) = 2nk,’. Actually the same mid-parental system may be repeated many times, only in this case the possible correlation of fertility with the character under discussion must be guarded against. 406 ; Prof. Karl Pearson. and note that S(nk/)=N2/, S(nk,k,) == NE eae where oy) is the standard deviation of the offspring, N is the total number of pairs of brothers or mid-parents of each order, and gq is as before the standard deviation of the group of gth mid-parents. Noticing that py .g = ca't’~?, we have if r be the correlation between brothers Nogr = S[S(a:)] = 28 (nk?) = 2Noi? S(o?p"+.29/°p0t'eai"'—0)), the sum now referring to all values of ¢ and q’ from 1 to «, q being unequal to q’, and q’ ~ q taken positive. Thus: y = 27°(P? + Boat Bica?+ Peat... + B®ca+ B+ Pica + Biea* + + B*ca? + B’ca+ Bo + Blica+ oe + Pca? + B8ca*? + Blea + PS+ eee ose sas oo 52 tele pee el ) Hence summing parallel to the diagonal : Be 2caB \ Bas Sea ke.) es acs i as eae Ce ee oe xvii). 2 me Cane by (x11), (xi), and (xvii). Let us evaluate this on Mr. Galton’s law and on by hypothesis of a 10 per cent. tax.* On the first hypothesis B = —-, and iD a ae ial? y = 1, hencer = 0°4. On the second hypothesis (p. 403) 6 = 0°39284, a = 0°74639, and y = 0'9; hence r = 0°4402. We can also obtain less accurate values of fraternal correlation in other ways. Suppose two brothers to be considered as sons of one mid-parent k, only. In this case we must take 0°6 for the regression (see the table, p. 403), or @, = 06h, +2’, to = 06h, +2”, and as before: S(e@2) = 2x (06)? x Sk), No,’r = 2X 0°36 x N=,’, Y ==)0'S6, * [The above value for fraternal correlation shows that y must be >0°6076; that a must be <1, only gives y >0°5469. | Mathematical Contributions to the Theory of Evolution. 407 If we suppose two individual parents with no assortative mating, we have Bee 20s") aia ie’ v niu, he ara 9) Soc op (xix), where 7; and 7, are the male and female parental correlations. With Galton’s law 7; = 7, = 0°3, and r again = 0°36. Assuming the value 7, = 7 = adopted by Mr. Galton in his ‘ Natural Inheritance’ (p. 133) for parental regression, the fraternal regression deduced from this ought to have been 4 = 0°44, and not 0°67 as obtained by Mr. Galton.* The mean of the sister-sister, brother-brother, brother-sister correlations that I found in 1895,} duly weighted for the number of pairs in each case, is exactly 0°4000. The value as it might have been @ priori predicted from Galton’s law = 0:4000, with a rise to 0°4402, if we “‘ tax” up to 10 per cent. I conclude therefore that this law of ancestral heredity is at least to a first approximation in agreement as complete as could possibly be expected with the facts we as yet know as to collateral heredity. It confirms the view I took in 1895, that fraternal heredity cannot be taken greater than 0°5. I think the high value (about 0°6) obtained from Mr. Galton’s “special data”’ must be explained by my suggested causet (a) 2.e., unconscious selection of approximately equal heights in brothers who join Volunteer regiments; for the explanation (6) is taken away if we accept Galton’s law without a modified y. (11) Turning now to the inheritance of cousins, we notice that their regression may be represented by ae | ; 1 Seg TR ge Ji Ly = zh t+ihky eeee +ih, + 7h +2, ae oa ’ " at B= zhy' + ik cece +rth, '4+thy' +a". Here h, and h," are children of the same parents and have fraternal correlation; h, and h,” are their other parents, and without a double cousin marriage have no correlation with each other, or neglecting sexual selection with h, or hy’; ky is the mid-parental system§ of h,, and therefore of h,'; k)' and: k)’ the mid-parental systems of fh, and h,'", and accordingly, if there be no in-and-in breeding, uncorrelated with each other or with kp. Summing first for the array corresponding to fy, hy", S (a2) a nt 335 hyhy"’ +76 ko(ha + hy") +ieho}, * Mr. Galton took r = 27; this is part of what, I think, the erroneous theory of regression developed in ‘ Natural Inheritance,’ a theory which is inconsistent with the law of ancestral heredity given in the same work. fT ‘Phil. Trans.’ A, vol. 187, p. 281. f ‘Phil. Trans.,’ A, vol. 187, p. 284. § This means that &) = 4h.+4h3+4h,+ .... where #q is the common mid-gth parent of the two cousins. 408 Prof. Karl Pearson. where n is the number of pairs of cousins corresponding to hy, hy". The factor 2 (see footnote, p. 404) does not occur here, as the cousins form parts of separate, and not identical, arrays. Now let us sum for all possible mid-parental systems, then if r be the correlation of cousins and N the total number of cousin pairs : No,*r' = S[S (aa) |] = =, {S(nhhy") +8 rk( th") ]+8(nk,?) }. But S(nh,hi") = product moment for pairs of brothers = Na,’r. S(nk,?) = No’r, by what precedes. S[nko(i1+h')} is exactly the same as the sum of all offspring with the mid-parental system of ancestry beyond, since /, is not to be equal to i”, = 28[ nh (4h.+¢hst+thy+ ....)] for all values of hy. — 2N ($pi¢0 21+ 4p2¢022+ tpso023+ coos y 5 = 2No,"(5c2’+ ¢oa'+4ca°+ ....), » ECa” ' = 2Na,’ ii = 0'4 x Na,” Thus Nor" => wg No,” (2r +0°4), and 7 = 0075. Mr. Galton’s value is 2, = 0:074 (‘ Natural Inheritance,’ p. 133). Had we, however, applied his method correctly, considering cousins as the offspring of brothers, and adopted the value 0°3 given by his law of ancestral heredity for parent and offspring, we should have found 0:0360, instead of our present 0°075. Considering cousins as having two grandparents the same, we should have found 0°0450. Second Cousins—The correlated parts of their mid-parental systems are thatgehetseho, ri hy' +7) =F seh, where h, and h,’ are cousins, h, and h,’ brethren, and io = Skhatsthitehst+ eevee is the mid-parental system of h, and hy’. In order to work out the correlation, we shall clearly want that of h, and h,', or of nephew and uncle. Wo’ — kote’, give the correlated parts of the mid-parental systems. Mathematical Contributions to the Theory of Evolution. 409 Hence if r” be the uncle-nephew correlation and N the total number of pairs | No,’r ems 48 (nkohe) ++ £9 (nke’) == (Opa?) Noy’, since S(nheko) = ooN $pi2it+dp2te+ ....) = oN (See? +icat+ ....) pce aN aes Foe 9 l—3Ca == 022) X oN. | Thus 7’ =0'15, or is double the correlation of first cousins. Here, as throughout, the variations of all generations, in this case those of uncles and nephews, have been treated as equal. Returning to the correlation r'” of second cousins we have Nor’ = 2548 (nhyhy’) +258 (ahehe') + ps S(nk?) + $8 (nig he') +48 (oh Ta) +38 (nko F) +3 3S(nk nee) Evaluating each of these terms we have— S(ahhy') SN Sar 5 S(nhzhp') = ="Nr oo- 5 S(nko 2. = Nvro,?. S[2(hyhe' + hy'hz) |] = product moment of all uncles and nephews == IN7 o,. S| nko(y4+h;')| = product moment of all offspring and the mid- parental system of their grandfathers = 28(nk hf.) for all values of h, ( kee 1 a agar? )] = 28] nh, (4 4 = 2(3 1200 N+ trzo0N +i 70°N + eees ) ieee 3Ca = 2Noao'($ca' + toa +208 + sescos) = 2Nao" 5 gps ap a) ey X O'L. Similarly, S[ko(he+ hz") ] = 2No,” x 0'2 as before (p. 408). Thus finally: er = ger’ tysttret ter +4 Xx 02+ 7g X04) or | r'" = 0°0171875. More distant collateral relationships, which can be found in like manner, and may be needed for the case of in-and-in breeding, say 410 3 Prof. Karl Pearson. from single pairs, are given in the table below. This case, which offers some striking applications of Galton’s law, I postpone for the present. Collateral Heredity according to Galton’s Law. Relationship. Correlation. Pes EyenaTe: ave zee ieee lege 0 °4060 Uncle and nephew Shs adnlod onc 0 1500 Great uncle and nephew ....... 0 ‘0625 First cousins ...... Fae 0°0750 First cousins once nemoved.. 4 0 :0344 Second cousins. oe she 0 :0172 Second cousins once Gomovede 0 0082 Third cousins. 4 0 :0041 Had we regarded second cousins as grandchildren of brethren, we should have found 0:0090 instead of 0°0172 for example, showing the degree of approximation of the incomplete theory. (12) On Cross Heredity—In my memoir on heredity cf 1895, I have defined cross heredity as the correlation between different organs in any two relations.* If we consider Galton’s law of ancestral heredity to be applicable to the inheritance of any character or quality whatsoever, then we can obtain from it a solution of the whole problem of cross heredity. This solution seems so simple and plausible that it deserves careful consideration, and I hope shortly to be able to test it by the measurements in my possession. Let A and B be any two relatives; 1 and 3 represent any two organs in A, 2 and 4 the same organs in B. Now suppose we investigate the manner in which the index 1 to 3 is inherited by B, 1.¢., let us find the correlation between the indices lto3and 2 to 4. Let p be the coefficient of heredity between the degrees of blood A and B, and suppose it by Galton’s law to take the same value for all qualities and characters, then 7 will be the correlation not only between 1 and 2, and 3 and 4, but also between the indices 1 to3and 2 to 4, The value of this correlation was given by me in ‘ Roy. Soc. Proc.,’ vol. 60, p. 493, Equation iv, and is 91QV Vq— 7 4V1V4— Vo3VQV3 + 131V3V4 J VE V3?— 2730 V3 o/ Vo + VG? — 2 oyVV4 P= where 1, V2, v3, vs are the coefficients of variation of the four organs, and the r’s are their coefficients of correlation. Now if there be no secular selection v, = v2, v3 = V4, and 713 = 121; * ¢Phil. Trans.,’ A, vol. 187, p. 259. Mathematical Contributions to the Theory of Evolution. 411 further, by Galton’s law, 72: = 73: = p, for both are coefficients of direct heredity. Hence I a te (v, i v3") p —2U\v3 ——_—— 2 p Si v; + v3" — 2v,0,K r] where R is the organic correlation between the two organs in the same individual. Thus it follows at once that 4(rius +723) =p x d 5 Or the mean of the two coefficients of cross heredity is the product of the coefficient of direct heredity into the correlation of the two organs in the same individual. Now in all cases of interchangeable relation- ship, z.e., brother and brother, or cousin and cousin, 74 = 73, and it is highly probable that this is also true where the relationship is not interchangeable, e.g., parent and offspring.* Thus we reach the exceedingly simple rule for cross heredity. Multiply the coefficient of direct heredity by the coefficient of organic correlation, and we have the coefficient of cross heredity. For example, the organic correlation between femur and humerus is about 0°85 for Aino or French males. Hence we should expect to find the cross heredity between femur of parent and humerus of offspring to be about 0:3 0°85 = 0:25. Thus Galton’s law, even if it be not absolutely correct, will still serve as a useful standard to test the problems of cross heredity. (13) Conelusion.—The above illustrations of Galton’s law will suffice to prove the wide extent of its applications. If either that law, or its suggested modification, be substantially correct, they embrace the whole theory of heredity. They bring into one simple statement an immense range of facts, thus fulfilling the fundamental purpose of a great law of nature. It is true that there are difficulties which will have to be met, among which I would note two in par- ticular: _ (i) Galton’s law makes the amount of inheritance an absolute constant for each pair of relatives. It would thus appear not to be a — character of race or species, or one capable of modification by natural selection. This seems to me a priori to be improbable. I should imagine that greater or less inheritance of ancestral qualities might be a distinct advantage or disadvantage, and we should expect inheritance to be subject to the principle of evolution. This diffi- * For example, the correlation between the arm length of one brother and the stature of a second, must be equal to the correlation between the arm length of the second and the stature of the first. It is probable, but requires statistical confir- mation, that the correlation between stature of parent and arm length of offspring is equal to the correlation between arm length of parent and stature of offspring. 412 Mathematical Contributions to the Theory of Evolution. culty would be to some extent met by introducing the coefficient y, which I would propose to call the coefficient of heredity, and con- sider as capable of being modified with regard to both character and race. As such a law would cover Mr. Galton’s case, there does not seem any objection to using the more general formula, until it is found that the strength of heredity is the same for all characters and races. Of course it may well be argued that heredity is something prior to evolution, itself determining evolution, and not determined by it. If this be so, its absolute fixity for all organs and races ought to be capable of observational proof. (ii) For the inheritance of fertility in man from parent to offspring, Miss Alice Lee has recently worked out 6,000 male, and 4,000 female cases. The result shows that fertility is probably a heritable cha- racter, but the correlation between parent and offspring is scarcely one-tenth of that given by Galton’s law. The difficulties of any fairly exact determination of the amount of fertility inherited in man under the present artificial conditions are very great, but even allow- ing for these, I think we must assert that fertility is inherited in man, but in a degree very much less than Galton’s law would require. I hold, then, that, as far as our knowledge goes at present, we must be cautious about treating y as exactly equal to unity. That is a limiting value which certainly gives strikingly good results for a great deal of what is yet known, but we must wait at present for further determinations of hereditary influence, before the actual degree of approximation between law and nature can be appreciated. Hven with regard to such determinations, there must be no haste to assert that they actually do contradict Galton’s law. That law states the value of certain partial regression coefficients, the total regression coefficients that we have deduced from them are only correct on certain limiting hypotheses, the most important of which are the absence of reproductive selection, ¢.e., the negligible correlation of fertility with the inherited character, and the absence of sexual selection. I pro- pose to deal with the results of Galton’s law, when assortative mating is taken into account, especially in the’ case of in-and-in breeding, in another paper. At present I would mereiy state my opinion that, with all due reservations, it seems to me that the law of ancestral heredity is likely to prove one of the most brilliant of Mr. Galton’s discoveries; it is highly probable that it is the simple descriptive statement which brings into a single focus all the complex lines of hereditary influence. If Darwinian evolution be natural selection combined with heredity, then the single statement which embraces the whole field of heredity must prove almost as epoch- making to the biologist as the law of gravitation to the astronomer. Mathematical Contributions to the Theory of Evolution. 418 “ Mathematical Contributions to the Theory of Evolution. On the Inheritance of the Cephalic Index.” By Migs CicELy D. Fawcett, B.Sc., and KARL PEARSON, M.A., F.R.S., University College, London. Received January 27,—Read February 17, 1898. (1) The cephalic index, when used to test any theory of heredity, possesses many merits, and at the same time one or two defects. In the first place it is supposed to be a marked racial character, and therefore might be considered to be strongly inherited. In the next place it remains sensibly constant after two years of age; thus the strength of inheritance can be ascertained by measurements on young children, whose parents are more frequently alive than if we have to wait for measurement till the offspring are of adult age. Further, although the cephalic index requires a more trained hand to measure it than some other measurements on the living subject, the trained observer will always deduce sensibly the same results ;* on the other hand, stature measurements vary sensibly with the hour of the day, and with the observer. The need of a moderately trained observer is the chief defect of cephalic index measurements ; it hinders the rapid collection of numerous family measurements; the difficulty, further, of satisfactorily measuring the female head without some derangement of the toilet is a further hindrance.f ‘The merits of the cephalic index, however, as a test of heredity far surpass its demerits. A well-organised measurement of the cephalic index in pairs of relatives would probably give the best results available for the laws of inheritance. The cephalic index measured on the living head is of course not so satisfactory as that measured on the skull, but the latter may be considered, even with the aid of Rontgen rays, as at present quite out of the question. The following paper has been worked out, not on very good material or on material collected with the present end in view, but on the only — material that seemed at present available. It suffices to justify the view that the inheritance of the cephalic index offers a most satisfac- tory method of testing the laws of heredity. (2) Owing to the kindness of Mr. Francis Galton the Department of Applied Mathematics in University College, London, was placed ia communication with Dr. Franz Boas, of the American Museum of Natural History, who is well known for his elaborate system of * This has been tested by frequent measurements of the same heads. + The recent establishment of an anthropometric laboratory at Newnham College will, it may be hoped, remove the difficulty about head measurements on female students felt by the Cambridge Anthropometric Committee. 414 Miss C. D. Fawcett and Prof. Karl Pearson. measurements on North American Indians. With extreme kindness Dr. Boas* at once forwarded to England upwards of 1000 sheets of measurements on comparable Indian tribes. These tribes, however, contain extremely mixed blood. In the fewest cases were pure Indian ancestors noted; one of the grandparents at least exhibited as a rule European blood—English, Dutch, French, Irish, &e. Dr. Boas himself writes :— ‘“T could not give you any series that was sufficiently extensive and embraced pure Indians only, because among these tribes the determination of relationships offers peculiar difficulties. 1 am afraid that your results may also bring out the looseness of family relations. I should not be surprised if the relation between father and child were much lower than that between mother and child, because often another person is actually the father of the child.” Dr. Boas’s last surmise is amply verified ; it will be found from the table below that the coefficient of heredity between father and son is abnormally small, while that between father and daughter is actually less than the probable error of this series of measurements! If we put upon one side any purely hypothetical supposition that illegiti- mate births are more likely to be female than male, there would seem reason to suppose some native custom by which it is held less discreditable to pass off a daughter than a son upon the titular husband. It may be asked whether, if the racial mixture is so great and the paternity so cbscure, it was worth while to undertake the lengthy arithmetict required to determine the hereditary correla- tions. The answer is threefold: (a) if Galton’s law of ancestra! heredity be correct, inheritance is not a racial character but a general law of living forms, and racial mixtures will not influence the result ; (b) the results show that obscure paternity does not prevent good values being found for other relationships; in fact, the fulfilment of Dr. Boas’s surmise is in itself not without value, as showing how well our algebraic theory fits itself to the facts ; it might almost be said to provide a scientific measure of the conjugal fidelity of a race; (c) it is always worth while to undertake an investigation on the best material availabie, even if it be poor material for this pur- pose, for it emphasizes the need of new and more elaborate obser- vations. (3) It will be seen from the tabie that it has only been possible to determine the coefficient of heredity for small series, varying from 80 to 143 pairs of the seven relationships, four corresponding to the first degree of direct kinship and three to the first degree of collateral * Itis difficult to sufficiently emphasize the disinterested service to science of men who do not “ monopolise” their anthropometric measurements. + We have to thank Mr. Leslie Bramley Moore for much aid in extracting the head measurements from the slips and calculating cephalic indices. 415 ,corenbs UBvOotTT JO AOLL9O 9 AO) MOMVIANp-pxrVepueys = ‘a s! % (Se Se Sy et ee = 6660. OF | 118. OF 690. F OI. OF FB 18 08 8 O00F- 0 OSPP- 0 FLG0. OF GSSF- 0 { Pe pneoeend 1. OF OL: 28 08 Ss eS OLIO. OF CFT. OF 889-S 02-08: TS CPT = ees res Bong Dee ELS 6E1- OF OGF- ¢ 03-0F8918 | eT = > E20. OF ZST-OFS94-8 1Z- OF ZF: 18 al S 000F- 0 €168-0 06F0- OF L8LE-0 { OST. OF ZS9- & GI. OFZ: 08 6EI fad By 620- OF SIZ. OF SELF Te. OF 9-18 ag ne ~~ 9 6460. OF COT. OF P29: & €Z. OF Ge: 18 FOL eo on0E-0 Si0En0 vg0-0F 96980 {| Trowoco.e 02-0F08-08 | TOT ~ 3 9160.07 ZST- OF 916. & 9Z. 0F 06: T8 got : 660. OF ChL. OF SEF S 02-0 gg. 18 Tt ~ 2 eee a re ‘£LOOT T, oANgRyg ‘xepuy oreydeg 8 3 «C'S “eo ON s “TLOTJBOALOD FO FUOLOTOO” Die en oe ee en ie ee ee ees See | aa ae ‘gon[eA Jo o[qey—xepuy orpeqdap jo soueytroyUy oem eo Pe own ere ee ew eevee ee ee oe S1998IS ooo Fo oe 6 eG oe te eee "UOT}R OT "* SLOqSIg. "* STOqSTG SLIGO SLOYIOLT SLoy{org, SLOySNecy ** SLOTIOTAL oo oe suog '* sLaTJOIL oeeove re oO Oe e10Vonecy Pee ee ee ee SLO IVA eeecrree ers oe eer oe suog eoeevese00 8868 8 SLO 4G 416 Mathematical Contributions to the Theory of Evolution. kinship. The probable errors are, as might be expected from such small series, large. Putting aside the paternal relationship, we are justified in drawing certain general conclusions, which may be thus summed up :— (a) The coefficients of heredity, as determined from the cephalic index, differ in all cases from those determined for stature by less than their probable error, and therefore by less than the probable error of their difference. The stature coefficients were obtained for the English middle classes.* We thus conclude that these results confirm Galton’s law, in so far as they tend to show that the strength of inheritancets not a character of race or organ. Cephalic index is clearly not more strongly inherited than stature. Its variability is also very much that of stature. It is accordingly difficult to see why it should be considered as peculiarly a racial character. (b) The divergences between the observed values for the coefii- cients of inheritance for the cephalic index, and the theoretical values obtained on the basis of Galton’s law of ancestral heredity, are ereater than the divergences between the former and the coefficients for stature.t| They do not, however, exceed the limits of errors of observation. In the case of mothers and sons the divergence is very slightly above the probable error; the observed and theoretical values are identical in the case of mothers and daughters; they are less than the probable error for brothers and brothers and only slightly larger than it for brothers and sisters; for sisters and sisters the divergence is about one and a half times the probable error. The mean weighted values of the coefficients for direct and collateral kinship are 0°3366 and 0°4004, the former differing by less than half its probable error from the theoretical value 0°3000, and the latier sensibly identical with its. theoretical value, 0°4000. We conclude, therefore, that Galton’s law of ancestral heredity gives values for the inheritance within the limits of the prebabie errors of observation. But, (c) As in the case of stature there is, on the whole, a tendency of the coefficients for cephalic index to be somewhat greater than their values as given by Galton’s law. It is therefore reasonable to sup- pose that the heredity constant y (introduced in the foregoing paper “On the Law of Ancestral Heredity’’) is not, as Mr. Galton takes it, unity, but has some slightly less value. Other conclusions which may be drawn from the above table are: (2) Among Indians of mixed blood the women are more brachy- * ‘Phil. Trans.,’ A, vol. 187, pp. 270 and 281. + It is to be noted, that, putting paternity aside, the order of relative magnitude of the coefficients of heredity is precisely the same for both cephalic mdex and stature. Comparison of Oxygen with Helium Stars, Sc. 417 cephalic and more variable than the men. This isin accordance with the general conclusion reached in a paper on “ Variation in Man and Woman,” * namely: “The lower races give us results in sensible accordance with those we have drawn from the data for ancient civilisations, namely, the women are on the whole more brachycephalic and slightly more _ variable than the men.” (e) The younger generation is more brachycephalic and more variable than its parentage. The whole of this difference can hardly be due to any change of shape of the skull with old age, for the majority of parents had in this case not passed the prime of life. It may be due to (i) a corre- lation between dolichocephaly and fertility or between dolichocephaly and philogamy, or (ii) more probably to the action of natural selection (results obtained, but not yet published, by the present writers show a correlation between physique and cephalic index), or (iii) to a greater or less admixture of white blood in the younger generation. (f) Parents of sons are significantly less variable than parents of daughters. This is in accordance with the result previously ob- tained that mediocre fathers are likely to have sons,} but disagrees with the result for stature—based on a far smaller probability—that mediocre mothers are likely to have daughters. The conclusions of this paper, while appearing to the writers. of interest, are to be taken, in the first place, as suggestions for much larger series of measurements and for new lines of investigation. “ Comparison of Oxygen with the Extra Lines in the Spectra of the Helium Stars, 8 Crucis, &c.; also Summary of the Spectra of Southern Stars to the 34 Magnitude and their Distribution.” By FRANK McCuLEAN, F.R.S. Received January 12,—Read February 3, 1898. | PLATE 6. | In a previous paper read before the Society on April 8, 1897, I suggested that the special lines present in spectra of the first division of helium stars (Type I, Divison 1a) might possibly be due to oxygen. These stars are associated by their position and distribu- tion with the gaseous nebule, and some of the lines in their spectra correspond with bright lines observed by Campbell in nebule. The suggestion from this was that these stars are in the first stage of stellar development from gaseous nebule. * Pearson, ‘The Chances of Death,’ vol. 1, p. 370. + ‘Phil. Trans.,’ A, vol. 187, p. 274. Woe. LX. Qn 418 Mr. F. McClean. Comparison of Oxygen with the The special lines referred to are the extra lines which distinguish these spectra from those of the remaining helium stars of Divi- sion Ib. The indications in the spectra of the northern stars that these extra lines are due to oxygen are slight, as the lines at best are indistinct. Among the southern stars, however, there are several in the spectra of which these lines are better defined, and there is one, viz., 8 Crucis, in which they are very fairly defined. The following stellar spectra are mounted on the fy 3 ee plate, viz., « Orionis, B Scorpii, @ Canis Majoris, 8 Centauri, and 8 Crucis. These photographs are intended to show the gradual im- provement in the definition of the extra lines, between « Orionis and f Crucis, and to indicate their identity of origin throughout. The extra lines in the spectrum of 6 Crucis are singled out by comparison with another helium star, viz., « Argus, of Division I), in which the extra lines do not appear. ‘The lines are drawn out by themselves below the spectrum of 6 Crucis. They are then com- © pared directly by juxtaposition with a drawing of the spectrum of oxygen as tabulated in the spectrum of air by Neovius (Stockholm, 1891, and Appendix EH, 1894, of ‘ Watts’s Index’). This comparison shows a close correspondence in the grouping of the extra lines with the spectrum of oxygen. The most remarkable correspondence is in the case of the large group on either side of Ho. A slight shift of about a tenth metre is required to bring the groups into identical positions. However, the close similarity of the whole grouping of the two spectra as they appear on the plate admits of little doubt that the extra lines actually constitute the spectrum of oxygen. If this be established the spectrum of the first division of helium stars would be due to hydrogen, helium, and oxygen. The scale attached to the spectra is based on standard lines that can be identified with certainty in the stellar spectra. It is inter- polated between the standard lines. Its position in relation to the spectra is determined by the hydrogen lines. The wave-lengths em- ployed are in accordance with Angstrém’s scale. On the original negatives the distance between (H) and (F) measures about 1 inch. The negatives are enlarged about eight and a half times. It is difficult to fix the position of the lines—and especially of the hydrogen lines—on these enlargements with sufi- cient accuracy. A further correction than this would account for is however required in order to reduce the two spectra to exact coinci- dence. I believe it should be sought to some extent in a re- examination of the adopted wave-lengths of the hydrogen and of the oxygen spectra. ‘ The spectrum of y Argus is given on the plate in order to identify extra Lines in the Spectra of the Helium Stars, Fc. 419 it as a helium star. It contains two crucial lines of helium. The Wolf-Rayet stars, of which it is the principal example, are thus classified as helium stars. There are also some coincidences between the bright lines of y Argus and the spectrum of oxygen, which suggest a possible connection. The spectrum of » Centauri is also given as a bright line helium star. The bright lines in this case are due to hydrogen, and the spectrum resembles that of y Cassiopeiz. The spectrum of 6 Centauri is similar. I take this opportunity of presenting a summary of the spectra of 116 stars to the 35 magnitude in the Southern Hemisphere. They were photographed between May and October last by means of my own object-glass prism, mounted in front of the Cape astrographic telescope. This instrument, which is similar to my own telescope at Rusthall, with which the spectra of the northern stars were photo- graphed, was kindly placed at my disposal by H.M. Astronomer, Dr. Gill. It may be a little time before the actual photographs of the stellar spectra are ready for presentation, and meanwhile the results are of interest. In my previous paper I divided the sphere into eight equal areas consisting or two galactic equatorial areas and two galactic polar areas, situated on either side of the yalactic equator. The northern stars already given occupy the upper or northerly lateral areas A, B, C, and D, also the southerly area AA. The southern stars now given occupy the lower or southerly lateral areas BB, CC, and DD. Their photographic spectra are distributed into these areas, and are classified on the same system as in the previous paper. The table of distribution for the whole sphere by areas and classes is given below. There are in all 89 helium stars (Division I), distributed 71 in the galactic zones and 18 in the galactic polar areas, the areas being equal. There are 29 in the upper galactic zone (B and BB), and 42 in the lower galactic zone (C and CC). There are 9 in the upper polar areas (A and AA), and 9 in the lower polar areas (D and DD). There are 23 in the northerly halves of the two galactic zones (B and C) and 48 in the southerly halves (BB and CC). The 81 stars in Division II, the Sirian stars, and Division III, the Procyon stars (which along with Division I constitute Secchi’s Type I) are rather irregularly distributed throughout the sphere. There are 40 in the galactic zones and 41 in the galactic polar areas. There are 18 in the upper galactic zone (B and BB) and 22 in the lower (C and CC). There are 29 in the upper polar areas (A and AA) and 12 in the lower (D and DD). To the extent of the obser- vations there is no condensation of stars of Divisions II and III in the galactic zones as there is in the case of stars of Division I. The 106 stars in Divisions IV and V (II and III of Secchi’s types) 420 Mr. F. McClean. Comparison of Oxugen with the are fairly evenly distributed throughout the sphere. There are 52 in the galactic zones and 54 in the galactic polar areas. There are 22 in the upper galactic zone (B and BB) and 30 in the lower (C and CC). There are 27 in the upper polar areas (A and AA) and 27 in the lower (D and DD). The general distribution of the types of spectra throughout the sphere to the extent of the observations bears out generally the conclusion that stars with spectra of the more advanced types, in order of development, are evenly distributed in space. Also that stars with spectra more recent in order cf development are mostly congregated in the galactic zones. The helium stars of Division I are predominant in the Southern Hemisphere, being congregated in the lower or southerly halves of the galactic zones (BB and CC). They include 48 stars out of a total of 94 stars in those areas. They are also more closely congregated in the vicinity of the galaxy than is the case in the northerly halves of the galactic zones. In the contiguous constellations of Musca, Crux, Centaurus, Lupus, and Scorpio there are 27 helium stars out of a total of 36 stars included in the tables. (The distribution of the helium stars throughout the sphere was illustrated by two small hand charts, not reproduced, on which these stars are coloured red.) Appa- rently the region in which the first stage of stellar development is now most active lies in the southerly half of the galaxy. Table I. Photographic Stellar Spectra—Stars to Magnitude 33. Summary of Southern Stars—Regions BB, CC, and DD. Mag. | Div. | Area. Mag. Div. | Area. Aquila. Argo. aA 3°3 I (6) | CC a 2°5 LY, BB be 2°9 IV BB Ara. v 3°5 I (4) CC a 29 I (8) COC 3 3 °4: IV BB B 2°8 CC 7 ya | IV CC Y 366m! l(a) ace 0 | 3-2 III BB G 3°2 IV CC o 3°5 IV CC T 3°2 IV CC Argo. uv 3°4 III CC a 0°4 iII CC B 2°0 II CC || Canis Major. OY 3°0 I (@) CC | a —1°4 it CC 6 2°2 If CO B 2°0 I (a) CC E 2 IV CC 0) ee VE CC 4 2°5 I (0) CC E 1°5 I (a) CC ) 29 I (a) CC g 3°0 I (0) CO u 2°5 Til CC n 2°4, I(4) | CC kK 2°7 I (2) CO 0? 3°0 I (8) CC extra Lines in the Spectra of the Helium Stars, Jc. 421 Table I—continued. Mag. Div. | Area. Mag. Div. | Area. Capricornus. Libra. B 3°4 IV CC o (20) 3°2 Vv BB Centaurus. Lupus. a 0°7 IV CC a 2°6 I (a) | BB B 1:2 I (a) BB B 2°8 I (a) | BB y 2 °4 TE BB ¥ 3°2 I (a) | BB 8 2°8 | I(s) | BB 5 3-7 | I(a) | BB g 2°6 I (a) BB E 3°7 I (3) BB 4 27 | 1(6) | BB n 2°5 I (8) BB || Musca. 0 2°7 IV BB a 2°9 I (6) CC l 3°0 III BB B 3 °4 I (0) CC kK 3°3 I ( BB A 3°4 I (0) CC || Ophiuchus. Ke 3°4 I t BB 8 2°9 IV BB 2°8 I (a) BB Circinus. n 2°6 II BB a 3°5 Til CC 6 3°4 I (6) BB Kk 3°4 IV BB Columba. a. 2 I (0) CC || Pavo. B 2°9 CC a. 21 I (6) DD poe. B 3°3 Til DD a 13 | I(a) | BB , = sai ile aa |e B 1 er | I (a) BB picts Y 2°0 Vv BB } : B 3°3 IV. | DD Doradus. Y 34 IV DD a ae I (4) — Piscis Austr. Eridanus. * 1°3 i DD A : 2 10) BP Reticulum. “ Be ay oh Sagittarius. Grus. 7? 3°0 IV CC a 1°9 I (a) | DD 0 2°8 IV CC B 2°2 DD & 2°1 I (6) CC Y 3-0 | 1(8) | DD 2°9 ie pace é 3°5 II DD n 3°0 vV CC A 3°1 IV CC Hydrus. T BL FO) Oe a 2°9 III DD o 2°3 I 3 CC B 2°7 IV DD ? 3°3 I (0) CC Y 3°2 Vv DD Scorpio. Indus. a it vV BB a 3°1 BY: DD p’ 2°9 I (a) BB 6 2°5 I (a) BB Lepus. é 22 EV he Be a 2°7 III CC 0 2°1 III CO B 3°0 IV CC ve! 3°3 III CC i 3°3 IV CC kK 2°6 I (a) CC Ke 3°3 I (8) cc A 1% I (a) CC 422 Comparison of Oxygen with the Helium Stars, &c. | Mag Scorpio. pb 3°6 7 3-1 o 3°0 Tv jin) v 2°8 Serpens n | oc | Table I—continued. BET | Telescopium. a a 3°5 I (0) CC - Toucan. ; ro 2°8 IV DD | Trangulum. a 2°2 IV CC B 31 Iit CC 7 3°1 II CC Notr.—The magnitudes are taken from the ‘ Nautical Almanac’ (or from Gould). Table I. Summary Tables of Distribution of Gaseous Nebule and of Stellar Types. Stars to the 3; Magnitude. Stellar Types. Planetary nebule..... Extended nebule...... Total gaseous nebule .. Table No. 1. A. | B. | C. | D. | Lotal. | AA.| BB.| CC.| DD,| Total. 2| 8|.8| 2) (is) ) 2 eee des Vl 4|.8| 41 Ga) | 2 eee ere 3/7116] 6| G2)|'3 | ible te Nort.—Extracted from Table in Frost’s edition of ‘ Scheiner’s Astronomical Stellar Types. Davison © jp. ts ee ek es MEY ibcabak: SOE get eo tee Ha tf Gem Rete A. Spectroscopy.’ Table No. 2. B. | C. | D. | Total.| AA.| BB.| CC. | DD.} Total. iat ae ed eee eee eee ee eee CO 6|17| 3| (29) | 6] 23] 25} 6 | (60) 7 | ‘o4t3:| (90) | Sue ened lay $1 18| 4:| 27) |) 9.) eee | em) gs} 9/13] (44)| 9] 9]16] 9] (43) 2| 41's | (10) | Sue l(a) 31 38 | 26 |(180) | 30 | 38 | 56 | 22 |(146) k Orionis £ Scorpii | BP Can. Maj. 2 Centauri B Crucis y Argus Type 1, Div. 1b. 6 | | p Centauris . Type 1, Div. 1b. 7 | k Argus B Crucis: Type I, Div. la. 9 | | | Oxygen Cleveite Gas. Extra Lines 2.4 2.9 1.2 1.7 3.0 34 2.4 1.7 Sa ORO ae eel ae Se ee — : “McClean. Ac & = ee : i s —— - ] j S"s PHOTOGRAPHIC STELLAR SPECTRA, | Comparison of Oxygen with the Extra Lines in the Spectra of the Helium Stars, B Crucis, &c. (Div. Ia). — i Typed, Div. la. } k Orionis 24 Nn 2 Scorpii 2.) 3 B tan. Maj. 2.0 . 4 2B Contanri 1.2 5 3 (rueis 17 Type 1, Div. Ih. 6 y Argus 3.0) zo =S. x | ell r | | foe ; Ini sesezies K l Loan | | aso | Cloveito Gas. Typo 1, Div. th. 7 p Contauris 34 hk Argus 27 B Crucis 17 Type l, Div. la. 9 Extra Lines ni} | i| Oxygen Tatensthies " Me Kates HN, A ato, ie See oF Ware benath Meeting for Discussion. | 423 Table No. 3. ‘4. | 8./ 0. |. | rota.| 44.) BB ICC. DD! Total. Stellar Types. Division I ......... Sesu eel) 3 5 (29): | 6: |: 28" |x25 (60) ~ Peand 1D .-:.|97 | 15) 8 |. 7) (47) |-12 | 8 )14). 8 | G4) Bey end V..... 15 | 10 | 13| 16 | (54) | 12 | 12 | 17] 11] (52) eS ee ee ee 35 | 31 | 38 | 26 |(130) | 30 | 38 | 56 | 22 |(146) Table No. 4. Aand | Band C and D and Total | AA. BB. CC. DD. ote Stellar Types. ———— | LS 9 29 42 9 (89) Mettand Ir...) 29 18 22 12 (81) ety and Vi -..s.| 27 22 30 27 (106) a a 65 69 94 | 48 | (276) February 24, 1898. Sir JOHN EVANS, K.C.B., D.C.L., LL.D., Treasurer, in the Chair. Meeting for Discussion. Subject :—The Scientific Advantages of an Antarctic Expedition. The Discussion was opened with a communication by Dr. John Murray, and the following gentlemen contributed remarks :—The Duke of Argyll, Sir J. D. Hooker, Dr. Nansen, Dr. G. Neumayer of Hamburg, Sir Clements Markham, Dr. A. Buchan, Sir A. Geikie, Dr. Sclater, Professor D’Arcy Thompson, Admiral Sir W. J. L. Wharton. | i) el VOL. LXIL. APA Dr. J. Murray. “The Scientific Advantages of an Antarctic Expedition.” By JOHN Murray, D.Sc., LL.D., Ph.D., F.R.S. Received January 25,—Read February 24, 1898. . : From a scientific point of view the advantages to be derived from a well-equipped and well-directed expedition to the Antarctic would, at the present time, be manifold. Hvery department of natural knowledge would be enriched by systematic observations as to the order in which phenomena coexist and follow each other in regions of the earth’s surface about which we know very little or are wholly ignorant. It is one of the great objects of science to collect observations of the kind here indicated, and it may be safely said that without them we can never arrive at a right understanding of the phenomena by which we are surrounded, even in the habitable parts of the globe. Before considering the various orders of phenomena, concerning which fuller information is urgently desired, it may be well to point out a fundamental topographical difference between the Arctic and Antarctic. Inthe northern hemisphere there is a polar sea almost completely surrounded by continental land, and continental conditions for the most part prevail. In the southern hemisphere, on the other _ hand, there is almost certainly a continent at the South Pole, which is completely surrounded by the ocean, and, in those latitudes, the most simple and extended oceanic conditions on the surface of the globe are encountered. The Atmosphere. One of the most remarkable features in the meteorology of the globe is the low atmospheric pressure at all seasons in the southern hemisphere south of latitude 45° S., with the accompanying strong westerly and north-westerly winds, large rain- and snow-fall, all round the South Polar regions. The mean pressure seems to be less than 29 inches, which is much lower than in similar latitudes in the northern hemisphere. Some meteorologists hold that this vast cyclonic system and low-pressure area continues south as far as the pole, the more southerly parts being traversed by secondary cyclones. There are, however, many indications that the extreme South Polar area is occupied by a vast anticyclone, out of which winds blow towards the girdle of low pressure outside the ice-bound region. In support of this view it is pointed out that Ross’s barometric observa- tions indicate a gradual rise in the pressure south oz the latitude of 75° §., and all Antarctic voyagers agree that when near the ice the The Scientific Advantages of an Antarctic Expedition, 425 majority of the winds are from the south and south-east, and bring clear weather with fall of temperature, while northerly winds bring thick fogs with rise of temperature. - All our knowledge of the meteorological conditions of the Ant- arctic is limited to a few observations during the midsummer months, and these indicate that the temperature of the snow- covered Antarctic continent is even at that time much lower than that of the surrounding sea. The anticyclonic area at the South Pole appears therefore to be permanent, and when in winter the sea- ice is for the most part continuous and extends far to the north, the anticyclonic area has most probably a much wider extension than in summer. This is indicated by the south-easterly winds which at times blow towards the southern point of the American continent in June and July. | All observations in high southern latitudes indicate an extremely low summer temperature. In winter we have no direct observations. The mean of Ross’s air temperatures south of latitude 63° S. was 28°74° F., which is about the freezing point of sea-water, and his maximum temperature was 43°5° F. Both Wilkes and D’Urville observed pools of fresh water on several icebergs, and, when sailing along the ice barrier, Ross saw “gigantic icicles depending from every projecting point of its perpendicular cliffs,”* so it is probable that extensive melting sometimes takes place. In the latitude of the Antarctic circle the air is frequently at or near the point of saturation, and precipitation takes place in the form of rain, sleet, snow, or hail. Most of the observations near the ice-covered land show, however, a much drier atmosphere, and in all probability precipitation over the Antarctic continent takes place in the form of fine snow crystals, such as is recorded in the interior of Greenland. ' There would appear, then, to be good reasons for believing that the region of the South Pole is covered by what may be regarded prac- tically as a great permanent anticyclone, with a much wider exten- sion in winter than in summer. Ii is most likely that the prevailing winds blow out from the pole all the year round towards the sur- rounding sea, as in the case of Greenland, but, unlike Greenland, this area is probably seldom traversed by cyclonic disturbances. But what has been stated only shows how little real knowledge we possess concerning the atmospheric conditions of high southeru latitudes. It is certain, however, that even two years’ systematic observations within these regions would be of the utmost value for the future of meteorological science. * Ross, ‘ Antarctic Voyage,’ vol. 1, p. 237. rite 426 Dr. J. Murray. Antarctic Ice. From many points of view it would be important to learn some- thing about the condition and distribution of Antarctic sea-ice during the winter months, and especially about the position and motions of the huge table-shaped icebergs at this and other seasons of the year. These flat-topped icebergs, with a thickness of 1200 or 1500 feet, with their stratification and their perpendicular cliffs, which rise 150 or 200 feet above and sink 1100 or 1400 feet below the level of the sea, form the most striking peculiarity of the Antarctic Ocean. Their form and structure seem clearly to indicate that they were formed on an extended land surface, and have been pushed out over low-lying coasts into the sea. Ross sailed for 300 miles along the face of a great ice-barrier from 150 to 200 feet in height, off which he obtained depths of 1800 and 2400 feet. This was evidently the sea-front of a great creeping glacier or ice-cap just then in the condition to give birth to the table-shaped icebergs, miles in length, which have been described by every Antarctic voyager. All Antarctic land is not, however, surrounded by such inacces- sible cliffs of ice, for along the seaward faces of the great mountain ranges of Victoria Land the ice and snow which descend to the sea apparently form cliffs not higher than 10 to 20 feet, and in 1895 Kris- tensen and Borchgrevink landed at Cape Adare on a pebbly beach, occupied by a penguin rookery, without encountering any land-ice descending to the sea. Where a penguin rookery is situated, we may be quite sure that there is occasionally open water for a considerable portion of the year, and that consequently landing might be effected . without much difficulty or delay, and further that a party, once landed, might with safety winter at such a spot, where the penguins would furnish an abundant supply of food and fuel. A properly equipped party of observers situated at a point like this on the Antarctic continent for one or two winters might carry out a most valuable series of scientific observations, make successful excursions. towards the interior, and bring back valuable information as to the probable thickness of the ice-cap, its temperature at different levels, its rate of accumulation, and its motions, concerning all which points. there is much difference of opinion among scientific men. Antarctic Land. Ts there an Antarctic continent? It has already been stated that the form and structure of the Antarctic icebergs indicate that they were built up on,.and had flowed over, an extended land surface. As these bergs are floated to the north and broken up in warmer lati- The Scientific Advantages of an Antarctic Expedition. 427 tudes they distribute over the floor of the ocean a large quantity of glaciated rock fragments and land detritus. These materials were dredged up by the “ Challenger” in considerable quantity, and they show that the rocks over which the Antarctic, land-ice moved were gneisses, granites, mica-schists, quartziferous diorites, grained quartzites, sandstones, limestones, and shales. These lithological types are distinctively indicative of continental land, and there can be no doubt about their having been transported from land situated towards the South Pole. D’Urville describes rocky islets off Adélie Land composed of granite and gneiss. Wilkes found on an iceberg, near the same place, boulders of red sandstone and basalt. Borchgrevink and Bull have brought back fragments of mica-schists and other continental rocks from Cape Adare. Dr. Donald brought back from Joinville Island a piece of red jasper or chert containing Radiolaria and Sponge spicules. Captain Larsen brought from Seymour Island pieces of fossil coniferous wood, and also fossil shells of Cucullea, Cytherea, Cyprina, Teredo, and Natica, having a close resemblance to species known to occur in lower Tertiary beds in Britain and Patagonia. ‘These fossil remains indicate in these areas a much warmer climate in past times. We are thus in possession of abundant indications that there is a wide extent of continental land within the ice-bound regions of the southern hemisphere. Té is not likely that any living land-fauna will be discovered on the Antarctic continent away from the penguin rookeries. Still, an Antarctic expedition will certainly throw much light on many geological problems. Fossil finds in high latitudes are always of special importance. The pieces of fossil wood from Seymour Island can hardly be the only relics of plant life that are likely to be met with in Tertiary and even older systems within the Antarctic. Ter- viary, Mesozoic, and Paleozoic forms are tolerably well developed in the Arctic regions, and the cccurrence of like forms in the Antarctic regions might be expected to suggest much as to former geographi- cal changes, such as the extension of Antarctica towards the north, and its connexion with, or isolation from, the northern continents, and also as to former climatic changes, such as the presence in pre-Tertiary times of a uniform temperature in the waters of the ocean all over the surface of the globe. Magnetic and Pendulum Observations, Geodetic Measurements, Tides, and Currents. In any Antarctic expedition magnetic observations would, of | course, form an essential part of the work to be undertaken, and the importance of such observations has been frequently dwelt upon by eminent physicists and navigators. Should a party of competent 428 Dr. J. Murray. observers be stationed at Cape Adare for two years, pendulum obser- vations could be carried out there and at other points within the Antarctic, or even on the ice-cap and icebergs. It might be possible to measure a degree on the Antarctic continent or ice-cap, which would be a most useful thing to do. By watching the motions of the icebergs and ice from land at Cape Adare much would be learnt about oceanic currents, and our knowledge of the tides would be increased by a systematic series of tidal observations on the shores of the Antarctic continent, where we have at present no observa- tions. The series of scientific observations here mentioned, and others that might be indicated, would fill up many gaps in our knowledge of the physical conditions of these high southern latitudes. Depth of the Antarctic Ocean. In regard to the depth of the ocean immediately surrounding the Antarctic continent we have at present very meagre information, and one of the objects of an Antarctic expedition would be to supple- ment our knowledge by an extensive series of soundings in all directions throughout the Antarctic and Southern Oceans. It would in this way be possible, after a careful consideration of the depths and marine deposits, to trace out approximately the outlines of the Antarctic continent. At the present time we know that Ross obtained depths of 100 to 500 fathoms all over the great bank extending to the east of Victoria Land, and somewhat similar depths have been obtained extending for some distance to the east of Joinville Island. Wilkes sounded in depths of 500 and 800 fathoms © about 20 or 30 miles off Adélie Land. The depths found by the “Challenger”? in the neighbourhood of the Antarctic circle were from 1300 to 1800 fathoms, and further north the “Challenger ”’ soundings ranged from 1260 to 2600 fathoms. To the south-west of South Georgia, Ross paid out 4000 fathoms of line without reaching bottom. In the charts of depth which I have constructed I have always placed a deep sea in this position, for it appears to me that Ross, who knew very well how to take soundings, was not likely to have been mistaken in work of this kind. The few indications which we thus possess of the depth of the ocean in this part of the world seem to show that there is a gradual shoaling of the ocean from very deep water towards the Antarctic continent, and, so far as we yet know, either from soundings or tem- perature observations, there are no basins cut off from general oceanic circulation by barriers or ridges, similar to those found towards the Arctic. The Scientific Advantages of an Antarctic Expedition. 429 Deposits of the Antarctic Ocean. The deposits which have been obtained close to the Antarctic con- tinent consist of blue mud, containing glauconite, made up for the most part of detrital matters brought down from the land, but con- taining a considerable admixture of the remains of pelagic and other organisms. Further to the north there is a very pure diatom ooze, containing a considerable quantity of detrital matter from icebergs, and a few pelagic foraminifera. This deposit appears to form a zone right round the earth in these latitudes. Still further to the north the deposits pass in deep water, either into a Globigerina ooze, or into a red clay with manganese nodules, sharks’ teeth, ear-bones of whales, and the other materials characteristic of that deep-sea deposit. Since these views, however, as to the distribution of deep-sea deposits throughout these high southern latitudes, are founded upon relatively few samples, it cannot be doubted that further samples from different depths in the unexplored regions would yield most interesting information. Temperature of the Antarctic Ocean. The mean daily temperature of the surface waters of the Antarctic, as recorded by Ross, to the south of latitude 63° 8. in the summer months, varies from 27°3° to 33°6°, and the mean of all his obser- vations is 29°85°. As already stated, his mean for the air during the same period is somewhat lower, being 28°74°. In fact, all observa-— tions seem to show that the surface water is warmer than the air during the summer months. The “Challenger” observations of temperature beneath the surface indicate the presence of a stratum of colder water wedged. between warmer water at the surface, and warm water at the bottom. This wedge-shaped stratum of cold water extends through about 12° of latitude, the thin end terminating about latitude 53° S., its tem- perature varying from 28° at the southern thick end to 32°5° at the northern thin end, while the temperature of the overlying water. ranges from 29° in the south to 38° in the north, and that of the underlying water from 32° to 35°. This must be regarded as the distribution of temperature only during the summer, for it is 1m- probable that during the winter months there is a warmer surface, layer. , In the greater depths of the Antarctic, as far south as the Antarctic circle, the temperature of the water varies between 52° and - 30° F., and is not, therefore, very different from the temperature of the deepest bottom water of the tropical regions of the ocean. The presence of this relatively warm water in the deeper parts of the 430 Dr. J. Murray. Antarctic Ocean may be explained by a consideration of general oceanic circulation. The warm tropical waters which are driven southwards along the eastern coasts of South America, Africa, and Australia, into the great all-encircling Southern Ocean, there become cooled as they are driven to the east by the strong westerly winds. These waters, on account of their high salinity, can suffer much dilu- tion with Antarctic water, and still be denser than water from these higher latitudes at the same temperature. Here the density observa- tions and the sea-water gases indicate that the cold water found at the greater depths of the ocean probably leaves the surface and sinks towards the bottom in the Southern Ocean, between the lati- tudes of 45° and 56° 8. These deeper, but not necessarily bottom, layers are then drawn slowly northwards towards the tropics, to supply the deficiencies there produced by evaporation and south- ward-flowing surface currents, and these deeper layers of relatively warm water appear likewise to be slowly drawn southwards to the Antarctic area to supply the place of the ice-cold currents of surface water drifted to the north. This warm underlying water is evidently a potent factor in the melting and destruction of the huge table- topped icebergs of the southern hemisphere. While these views as to circulation appear to be well established, still a fuller examination of these waters is most desirable at different seasons of the year, with improved thermometers and sounding machines. Indeed, all deep-sea apparatus has been so much improved as a result of the ‘“‘ Challenger” explorations, that the labour of taking specific gravity and all other oceanographical observations has been very much lessened. Pelagic Life of the Antarctic Ocean. In the surface waters of the Antarctic there is a great abundance of diatoms and other marine algz. These floating banks or meadows form primarily not only the food of pelagic animals, but also the food of the abundant deep-sea life which covers the floor of the ocean in these south polar regions. Pelagic animals, such as cope- pods, amphipods, molluscs, and other marine organisms, are also very abundant, although species are fewer than in tropical waters. Some of these animals seem to be nearly, if not quite, identical with those found in high northern latitudes, and they have not been met with in the intervening tropical zones. The numerous species of shelled Pteropods, Foraminifera, Coccoliths and Rbkabdoliths, which exist in the tropical surface waters, gradually disappear as we approach the Antarctic circle, where the shelled Pteropods are represented by a small Limacina, and the Foraminifera by only two species of Globt- gevina, Which are apparently identical with those in the Arctic Ocean. A peculiarity of the tow-net gatherings made by the “ Challenger” The Scientific Advantages of an Antarctic Expedition. 481 Expedition in high southern latitudes, is the great rarity or absence of the pelagic larve of benthonic organisms, and in this respect they agree with similar collections from the cold waters of the Arctic seas. The absence of these larve from polar waters may be accounted for by the mode of development of benthonic organisms, to be referred to presently. It must be remembered that many of these pelagic organisms pass most of their lives in water of a temperature below 32° F., and it would be most interesting to learn more about their reproduction and general life-history. Benthos Life of the Antarctic Ocean. At present we have no information as to the shallow-water fauna of the Antarctic continent; but, judging from what we do know of the off-lying Antarctic islands, there are relatively few species in the shallow waters in depths less than 25 fathoms. On the other hand, life in the deeper waters appears to be exceptionally abundant. The total number of species of Metazoa collected by the *‘ Challenger ”’ at Kerguelen in depths less than 50 fathoms was about 130, and the number of additional species known from other sources from the shallow waters of the same island is 112, making altogether 242 species, or thirty species less than the number obtained in eight deep hauls with the trawl and dredge in the Kerguelen region of the Southern Ocean, in depths exceeding 1260 fathoms, in which eight hauls 272 species were obtained. Observations in other regions of the Great Southern Ocean, where there is a low mean annual tem- perature, also show that the marine fauna around the land in high southern latitudes appears to be very poor in species down to a depth of 25 fathoms, when compared with the number of species present at the mud-line about 100 fathoms, or even at depths of about 2 miles. In the year 1841 Sir James Clark Ross dredged off the Antarctic continent species which he recognised as the same as he had been in the habit of taking in equally high northern latitudes, and he sug- gested that they might have passed from the one pole to the other by way of the cold water of the deep sea. Subsequent researches show that, as with pelagic organisms, many of the bottom-living species are identical with, or closely allied to, those of the Arctic regions, and are not represented in the intermediate tropical areas. For instance, the most striking character of the shore-fish fauna of the Southern Ocean is the reappearance of types inhabiting the corre- sponding latitudes of the northern hemisphere, and not found in the intervening tropical zone. This interruption of continuity in the distribution of shore-fishes is exemplified by species as well as genera, and Dr. Giinther enumerates eleven species and twenty-nine genera as illustrating this method of distribution. 432 Dr. J. Murray. The following are the eleven species :—Chimera monstrosa, Galeus canis, Acanthias vulgaris, Acanthias Blainvillt, Rhina squatina, Zeus faber, Lophius piscatorius, Centriscus scolopax, Hngraulis encrasicholus, Clupea sprattus, Conger vulgaris. The genus by which the family Berycide is represented in the southern temperate zone (Trachichthys) is much more nearly allied to the northern than to the tropieal genera. ‘“‘ As in the northern tem- perate zone, so in the southern . . . . the variety of forms is much less than between the tropics. This is especially apparent on comparing the number of species constituting a genus. In this zone, genera composed of more than ten species are the exception, the majority having only from one to five.” . . . . “* Polyprion is one of those extraordinary instances in which a very specialised form occurs at almost opposite points of the globe, without having left a trace of its previous existence in, or of its passage through, the intermediate space.” Speaking of the shore-fishes of the Antarctic Ocean, Giimther says : “The general character of the fauna of Magelhaen’s Straits and Kerguelen’s Land is extremely similar to that of Iceland and Green- land. As in the Arctic fauna, Chondropterygians are scarce, and represented by Acanthias vulgaris and speciesof Raja . . . . As to Acanthopter ygians, Cataphracti, and Scorpenide are represented as in the Arctic fauna, two of the genera (Sebastes and Agonus) being identical. The Cottide are replaced by six genera of Tra- chinide, remarkably similar in form to Arctic types Gadoid fishes reappear, but are less developed; as usual they are accompanied by Mywine. The reappearance of so specialised a genus as Lycodes is most remarkable.’’* These statements with reference to shore-fishes might, with some modifications, be repeated concerning the distribution and character of all classes of marine invertebrates in high northern and high southern latitudes. The “Challenger” researches show that nearly * Ginther, ‘Study of Fishes,’ pp. 282—290. Edinburgh. 1880. + Ortmann, speaking of the Decapod Crustacea, says: “Nach dem Stande unserer jetzigen Kenntniss ist keine einzige bipolare Art bekannt” (Ortmann, ‘Zoologische Jahrbiicher,’ Abth. f. Syst., &c., Bd. ix, p. 585, 1896). Henderson, in his report on the “ Challenger’? Ancmuyra, in describing Lithodes murrayi from the Kerguelen region, says it ‘“‘is apparently most closely allied to Lithodes maia” (from the North Atlantic), “but the latter species is of large size, and the spines on the carapace are more numerous and more uniformly equal im size” (Hender- son, ‘Zool. Chall. Exp.,’ pt. 69, p. 44). Henderson writes me that these very slight differences were the only ones he could detect, and it seems evident that had the two specimens been taken from the same haul of the trawl, or from the same locality, they would never have been erected into two distinct species. Henderson writes me further that throughout the entire range of Crustacea there is no better illustration of bipolarity than that furnished by the Lithodide. For instances of The Scientific Advantages of an Antarctic Expedition. 433 250 species taken in high southern latitudes occur also in the northern hemisphere, but are not recorded from the tropical zone. Fifty-four species of sea-weed shave also been recorded as showing a similar distribution.* Bipolarity in the distribution of marine orgavisms is a fact, however much naturalists may differ as to its extent and the way in which it has originated. All those animals which secrete large quantities of carbonate of lime greatly predominate in the tropics, such as Corals, Decapod Crustacea, Lameilibranchs, and Gasteropods. On the other hand, those animals in which there is a feeble development of carbonate of lime structures predominate in cold polar waters, such as Hydroida, Holothurioidea, Annelida, Amphipoda, Isopoda, and Tunicata. This difference is in direct relation with the temperature of the water in which these organisms live, a much more rapid and abundant pre- cipitate of carbonate of lime being thrown down in warm than in cold water by ammonium carbonate, one of the waste products of organic activity. In the Southern and Sub-antarctic Ocean a large proportion of the Kehinoderms develop their young after a fashion which precludes the possibility of a pelagic larval stage. The young are reared within or upon the body of the parent, and have a kind of com- mensal connection with her till they are large enough to take care of themselves. A similar method of direct development has . been observed in eight or nine species of Echinoderms from the ' cold waters of the northern hemisphere. On the other hand, in temperate and tropical regions the development of a free-swimming larva is so entirely the rule that it is usually described as the normal habit of the Echinodermata. This similarity in the mode of develop- ment between Arctic and Antarctic Echinoderms (and the contrast to what takes place in the tropics) holds good also in other classes of Invertebrates, and probabiy accounts for the absence of free- swimming larve of benthonic animals in the surface gatherings in Arctic and Antarctic waters. What is urgently required with reference to the biological problems here indicated is a fuller knowledge of the facts, and it cannot be doubted that an Antarctic expedition would bring back collections and observations of the greatest interest to all naturalists and physiologists, and without such information it is impossible to discuss with success the present distribution of organisms over the surface of the globe, or to form a true conception of the antecedent conditions by which that distribution has been brought about. new species being made from purely geographical considerations, see ‘Summary of Results, ‘Challenger’’ Expedition,’ p. 1440-45. * Murray and Barton, ‘Phycological Memoirs of the British Museum,’ part iii. London. 1898. A34 ‘ The Duke of Argyll. Concluding Remarks. There are many directions in which an Antarctic Expedition would carry out important observations besides those already touched on in the foregoing statement. From the purely exploratory point of view much might be urged in favour of an Antarctic Expedition at an early date; for the further progress of scientific geography it is essential to have a more exact knowledge of the topography of the Antarctic regions. This would enable a more just conception of the volume relations of land and sea to be formed, and in connexion with pendulum observations some hints as to the density of the sub- oceanic crust and the depth of ice and snow on the Antarctic continent might be obtained. In case the above sketch may possibly have created the impression that we really know a great deal about the Antarctic regions, it is necessary to re-state that all the general conclusions that have been indicated are largely hypo- thetical, and to again urge the necessity for a wider and more solid base for generalisations. The results of a successful Antarctic Expedition would mark a great advance in the philosophy—apart from the mere facts—of terrestrial science. No thinking person doubts that the Antarctic will be explored. The only questions are: when? and by whom? I should like to see _ the work undertaken at once, and by the British Navy. I should like to see a sum of £150,000 inserted in the Estimates for the purpose. The Government may have sufficient grounds for declining to send forth such an expedition at the present time, but that is no reason why the scientific men of the country should not urge that the exploration of the Antarctic would lead to important additions to knowledge, and that, in the interests of science among Hnglish speaking peoples, the United Kingdom should take not only a large but a leading part in any such exploration. Riemarks by the Duke of Argyll. Scientific men generally feel, I think, that they do not need to give detailed reasons in connexion with particular subjects of inquiry, to justify their unanimous desire for an Antarctic Wxpedition. It is enough, surely, for them to point out the fact that a very large area of the surface of our small planet is still almost unknown to us. That it shouid be so seems almost a reproach to our civilisation. As to detailed reasons, it may almost be said with truth that there is hardly one of the physical sciences on which important light may not The Scientific Advantages of an Antarctic Expedition, 4385 be cast by Antarctic exploration. Oceanic circulation; meteorology ; magnetism ; distribution of animal and vegetable life, not only in the present but in the past; geology; mineralogy ; volcanic action under special conditions—all of these are subjects on which the phenomena of the Antarctic regions are sure to bear directly. If, however, I am asked to specify more particularly the question on which I look for invaluable evidence which can be got nowhere else, I must name, above all others, the most difficult questions involved in quaternary geology. Geologists are nearly all agreed that there has been, very recently, a glacial age—an age in which glacial conditions prevailed over the whole northern hemisphere to a much lower latitude than they prevail now. But geologists differ widely and fundamentally from each other as to the form which glacial agencies took during that period. In particular, many geologists believe in what they call an “ice sheet’’—that is to say, in the northern world having been covered by an enormous mass of ice several thousand feet thick, which, as they assert, “‘ flowed” over mountain areas as well as over plains, and filled up the bed of seas of a considerable depth. Other geologists disbelieve in this agency altogether. They deny that even if such a body of ice ever existed, it could possibly have moved in the way which the theory assumes. They affirm, also, that the facts connected with glaciated surfaces do not indicate the planing down by one universal sheet of enormous weight and pressure; but, on the contrary, the action or small and lighter bodies of ice, which have acted partially and un- equally on different surfaces differently exposed. We might have hoped that this controversy could be settled by the facts connected with the only enormous ice-sheet which exists in the Northern Hemisphere, viz., that which covers the great con- tinent of Greenland. But that ice-sheet, enormous though it be, does certainly not do what the ice-sheet of the Glacial Age is sup- posed to have done. That is to say, it does not flow out from Greenland, fill the adjacent seas, or override the opposite coasts, even in so narrow a Strait as Smith’s Sound. But this evidence is negative only. In the Antarctic continent we have reason to believe that there is a larger ice-sheet, and it certainly does protrude into the adjacent seas, not merely by sending out vast floating frag- ments, but in unbroken ice cliffs of great height. Now we want to know exactly under what conditions this protrusion takes place. Dr. Murray speaks of it as ‘“‘creeping”’ seawards—a more cautious. word than “ flowing.” But is it certain that it does even creep ?P May it not simply grow by accretion or aggregation till it reaches a depth of water so great as to break it off by floatation ? Does it, or does it not, carry detritus when no detritus has been dropped on its surface? Or does it pick up detritus from its owu bed? Or does 436 Sir J. D. Hooker. | it push foreign matter before it? Is the perfectly tabular form of the Antarctic icebergs compatible with any differential movement in the parent mass at all; or does it not indicate, on the contrary, a condition of immobility until their buoyancy lifts great fragments off? What is the condition of the rocks on which they rest? Is there any thrust. upon the mass from the mountain ranges on which the gathering ground lies? Oris the whole country one vast gather- ing ground from the continual] excess of precipitation over melting ? These questions, and a hundred others, have to be solved by Antarctic discovery; and until they are solved we cannot argue with security on the geological history of our own now temperate regions. The Antarctic continent is unquestionably the region of the earth in which glacial conditions are at their maximum, and therefore it is the region in which we must look fer all the informa- ation attainable towards, perhaps, the most difficult problem with which geological science has to deal. Remarks by Sir J. D. Hooker. Dr. Murray’s admirable summary of the scientific information obtainable by an organised exploration of the Antarctic regions leaves nothing further to be said under that head. I can only record the satisfaction with which I heard it, and my earnest hope that it will lead to action being taken by the Government in the direction indicated. Next to a consideration of the number and complexity of the objects to be attained by an Antarctic expedition, what dwells most in my imagination is the vast area of the unknown region which is to be the field for investigation—a region which in its full extension reaches from the latitude of 60° S. to the Seuthern Pole, and em- braces every degree of longitude. ‘I'his is a very considerable portion of the surface of the globe, and it is one that has been con- sidered to be for the most part inaccessible to man. I will therefore ask you to accompany the scientific explorer no further than to the threshold of the scenes of his labours, that you may see how soon and how urgently he is called upon to study some of those hitherto unsolved Antarctic problems that he will there encounter. In latitude 60° S. an open ocean girdles the globe without break of continuity. Proceeding southwards in it, probably before reach- ing the Artarctic circle, he encounters the floating ice fields which form a circumpolar girdle, known as “the Pack,” approximately con- centric with the oceanic, interrupted in one meridian only, that south of Cape Horn, by the northern prolongation of Graham’s Land. Pursuing his southward course in search of seas or lands beyond, after the novelty of bis position in the pack has worn off, he The Scientific Advantages of an Antarctic Expedition. 437 . asks where and how the components of these great fields of ice had their origin, how they arrived at and maintain their present. position, what are their rate of progress and courses, and what their influence on the surrounding atmosphere and ocean. I believe J am right in thinking that to none of these questions can a fuller answer be given than that they originated over extensive areas of open water in a higher latitude than they now occupy, that they are formed of frozen ocean water and snow, and that winds and currents have brought them to where we now find them. But of the position of the southern open waters, with the exception of the comparatively small area east of Victoria Land,* we know nothing, nor do we know anything of the relative amount of snow and ice of which they are composed, or of their age, or of the winds and currents that have carried them to a lower latitude. | The other great glacial feature of the Antarctic area is “the Barrier,” which Ross traced for 300 miles, in the 78th and 79th degrees of S. latitude, maintaining throughout the character of an inaccessible precipitous ice-cliff (the sea-front of a gigantic glacier) of 150—-200 feet in height. This stupendous glacier is no doubt one parent of the huge Mabie. topped ice-islands that infest the higher latitudes of the Southern Ocean; but, as in the case of the pack ice, we do not know where the barrier has its origin, or anything Sin WES about it, than that it in great part rests on the bottom of a compara- tively shallow ocean. It probably abuts upon land, possibly upon an Antarctic continent; but to prove this was impossible on the occa- sion of Ross’s visit, for the height of the crow’s nest above the surface of the sea was not sufficient to enable him to overlook the upper surface of the ice, nor do I see any other way of settling this im- portant point except by the use of a captive balloon—an appliance with which I hope any future expedition to the Antarctic regions will be supplied. There were several occasions on which such an implement might have been advantageously used by Ross when he was coasting along the barrier; and there were more when it would have greatly facilitated his navigation in the Pack. I have chosen the subject of the Antarctic ice as the theme for my acknowledgment of the honour you have done me in asking me to address this most important meeting, not only because it is one of the very first of the phenomena thes demand the study of the explorer, but because it is the dominant feature in Antarctic naviga- tion, where the ice is ever present, demanding, whether for being * T refer to the “ pancake’”’ ice, which in that area on several occasions formed with great rapidity around Koss’s ships, lat. 76° to 78° S., in February, 1842, and which arrested their progress. Such ice, augmerted by further freezing of the water and by snow, may be regarded as the genesis of fields that, when broken up by gales, are carried to the noi and contribute to the circumpolar pack. 438 Dr. G. Neumayer. penetrated or evaded, all the commander’s fortitude and skill and all his crew’s endurance. It may be expected that I should allude to those sections of Dr. Murray’s summary that refer to the Antarctic fauna and flora. They are wost important, for the South Polar Ocean swarms with animal and vegetable life. Large collections of these, taken both by the tow-net and by deep sea soundings, were. made by Sir J. Ross, who was an ardent naturalist, and threw away no opportunity of observ- ing and preserving; but unfortunately, with the exception of the Diatomaceee (which were investigated by Ehrenberg), very few of the results of his labour in this direction have been published. A better fate, I trust, awaits the treasures that the hoped-for expedition will bring back, for so prolific is that ocean that the naturalist need never be idle, no, not even for one of the twenty-four hours of daylight during a whole Antarctic summer, and I look to the results of a comparison of the oceanic life of the Arctic and Antarctic regions as the heralding of an epoch in the history of biology. Reemarks by Dr. G. Neumayer. With great pleasure I accepted the invitation to attend a discus- sion meeting on the importance, for the advancement of every branch cf science, of a scientific exploration of the Antarctic region. Re- gardless of the season and my advanced age, I hastened here to speak in the presence of so high a forum as the Royal Society of London, on the necessity of despatching as soon as possible an expedition towards the South Pole—an expedition cannot be dispensed with if we seriously desire the advancement of nearly every branch of human knowledge. It is fifty-five years ago since one of the greatest Arctic and Antarctic explorers ceased his work, so exceedingly well executed, in the Antarctic regions, and, since that time, it has never been taken up in any way comparable with that glorious scientific and nautical spirit manifested by Sir James Clark Ross. It is in view of this fact that we all look to the British nation as the one destined to carry on the exploration of the South Polar regions, and to assist this object as much as lay in my power, and to do homage to the memory of Ross, 1 could not fail to appear at this meeting. It is indeed a matter of great interest to examine the reason why so long a time has been allowed to elapse since the first great successes in the middle of this century. Undoubtedly political questions have interfered in an unusual manner so to retard progress in Ant- arctic inquiry, but it is not that alone; the cause is mainly that the thorough understanding of the importance of Antarctic research requires an unusual amount of knowledge, and not in one branch of science only, but in the whole complex of natural philosophy and The Scientific Advantages of an Antarctic Expedition. 439 natural science. The positive advantages to be derived from a renewal of Antarctic exploration appear but few at first glance, and can only be detected by a far-sighted man. The Royal Geographical Society has taken up the agitation for that renewal. The Sixth Inter. national Geographical Congress, two and a half years ago, devoted its influential exertions to the recommendation of it to the Govern- ment, not only of this country, but, indeed, of all civilised countries. But now that the Royal Society has taken the matter up (an example which will be followed, [ am convinced, by other academies of science), final success may be looked upon as matter of certainty. However much as has been done already to urge the importance of the scientific investigation of the South Polar regions, from as many points of view as there are branches of science cultivated by mankind, I consider it my duty, as a representative of geo- physical science, to add all the arguments in my power to those brought forward on the part of geologists, zoologists, botanists, and others. A gravitation survey in connection with a thorough geographical exploration of the Antarctic, is one of the most urgent requirements of the science of- our earth. There are no measurements of the gravitation constant within the Antarctic region, indeed they are very scarce in the southern hemisphere south of 30° south latitude. Such measurements are so closely connected with the theory of the figure of our earth, that it may well be asked how it can be con- sidered possible to achieve any advance in this respect, to arrive at any conclusive results in this all-important fundamental matter, without observations within the Antarctic region. It is impossible to foretell what effect an exact gravitation survey in that region might exert upon our views with regard to all physical and terrestrial constants, which depend on the radius of our earth. Apart from that consideration, we may hope for another important enlargement of our knowledge bearing upon the connection between terrestrial magnetism and gravity. Gravity observations have been so muck simplified of late by von Sterneck’s ingenious apparatus, that there is no serious difficulty in so multiplying gravity determinations within the Antarctic region, that we shall be quite justified in speaking of a “‘oravitation survey.” The all-important question of the distribu- tion of land within the Sonth Polar region is closely connected with it, and our leading authority on geodesy, Professor Helmert, justly lays great stress upon the observation of the force of gravity south of 60° south latitude. He induced the International Geodetic Permanent Commission to express it as their conviction that a gravitation survey within that region would be of the greatest benefit for higher geodetic theory (October, 1895). I have already alluded to the probable connection between gravity VOL. LXIL. 2K 440 Dr. G. Neumayer. and terrestrial magnetism. But apart from that, a magnetic survey of the Antarctic region is of the greatest importance from other points of view. As, since the time of Ross, no fresh observations on the values of the magnetic elements have been made, we are entirely ignorant of the values of the secular variation south of 50° latitude, data so much needed for the construction of reliabie magnetic charts. Of the situation of the southern magnetic pole, and of its motion ‘during the last fifty years, we are equally ignorant. The situation of the southern extremity of the earth’s magnetic axis and its motion throughout half a century are extremely important according to Gauss’s theoretical deductions. All computations, however intri- cate, must prove incomplete, and inadequately reward the immense amount of labour bestowed upon them, unless reliable values of the magnetic elements are first obtained, for which purpose a reduc- tion to a certain epoch by means of secular changes is indispensable. But not merely for theoretical purposes is such value to be put upon the knowledge of the magnetic character in the Antarctic, but also for the construction of reliable magnetic charts for use in naviga- tion. Certainly the studies of Carlheim-Gyllenskjéld and Van Bemmelen deserve great credit from a theoretical point of view, but they also cannot be carried to perfection unless we have a sound knowledge of the magnetic character of the South Polar regionr— at least as sound as that of the North Polar region. Jn the time of Ross, the magnetic Observatory at Hobart served as a safe basis for observations on terrestrial magnetism, and now the excel- lent Observatory at Melbourne may be turned to a similar account. Well as the mathematical theory of terrestrial magnetism has been developed, of the physical theory of that mysterious natural force, we are as yet in perfect ignorance. This deféct is certainly to some considerable degree caused by tke deficiency of our know- ledge in higher latitudes. It looks as if the magnetic character of the South Polar region were snch as would afford every facility for a sound investigation when brought into comparison with the magnetic conditions of the North Polar region. A glance at the map shows how entirely different is the distribution of the magnetic force (action) in the two polar regions. In the south there is to be noticed the interesting fact that the two foci of total intensity are both situated on the side towards the south of the Australian continent, and nearly on the same meridian. The magnetic action which makes itself manifest by magnetic storms or disturbances reaches its highest degree likewise south of the Australian continent, whereas to the south of South America these disturbances become very scarce and of a character in point of magnitude similar to those of the temperate zones. This was most strikingly proved by the observations in Orange Bay and Tee Sa Se ee eS ae, oe The Scientific Advantages of an Antarctic Expedition. 441 South Georgia during the period of International Observations in 1882-83. Of course, the magnetic South Pole and the situation of the foci above mentioned are in close connection with these facts, but the reason for this distribution remains unexplained. In agreement with this is the relative frequency of the South Polar lights within one and the same epoch. The map here shows the results of a discussion of all observations on southern lights, and, as may be perceived at a glance, the conformity with the fact just mentioned with regard to the region of maximum disturbance is striking. It may be mentioned, also, that a discussion of the mag- netic state of our earth for the epoch 1885 has yielded a curious fact (perhaps merely a coincidence), namely, that Dr. Schmidt, of Gotha, has calculated that part of the magnetic action of our earth which lies outside it, above the earth’s atmosphere probably, and has arrived at the conclusion that this part amounts to about 1/50th part of the entire potential. The curves which he constructed, based on this calculation, show likewise a close coincidence with the frequency of the southern lights. This requires to be further investigated, for it is perhaps a mere coincidence. The question of atmospheric electricity, yet under the shadow of a hypothesis of more or less probability, may yield, in connection with the matters touched upon, some results more definite than science has hitherto been able to divine. Under all these circumstances we perceive that there are a number of problems yet to be solved, which must stimu- late the scientific world to enter upon a close and conscientious examination of a region still enveloped in total darkness. The necessity of climatological researches within the Antarctic region has already been so much urged that it does not appear necessary for me to enter upon the matter. That it must remain impossible to arrive at exact climatological constants, so long as we do not know the winter temperature in that large area round the South Pole and bounded by the South Polar circle, is evident te every one conversant with the subject and does not require any further illustration. With regard to atmospheric pressure there is another curious fact which requires investigation, only to be carried on by observa- tions within the Antarctic. The excess of barometric pressure in 35° latitude over that in 60° latitude amounts in the northern hemi- sphere to 19:0 mm. and in the southern hemisphere to only 3'7 mm. Here is another problem worthy of examination. With the recital of these few facts, which suffice to prompt us to institute a vigorous examination of the South Polar regions the series is far from being exhausted. There is the question of the geoid-deformation, the phenomena of the tides, the structure of the ice and its drifting, especially the interesting fact of the appearance 2K 2 449 Sir C. Markham. of ice in lower southern latitudes, where it has never been observed since observations thereon are recorded. The resolution of the Sixth International Congress of Geographers, assembled here in August, 1895, according to which the present century ought not to be allowed to expire without the unveiling of the mysteries of the South Polar regions, ought to be carried into execu- tion by sending out an expedition to that end. All scientific institu- tions and societies have a well founded interest that such expedition should take place without further delay. Remarks by Sir Clements Markham. I need scarcely say how fully I concur in every word that has fallen from Dr. Murray on the subject of the scientific results, and more especially of the geographical results of an Antarctic Expe- dition. It is quite sufficient to point out the vast extent of the unknown area, and that no area of like extent on the surface of the earth ever failed to yield results of practical as well as of purely scientific value by its exploration. But there is much more to be said in the present instance, because the little that we do know of the Antarctic regions points unmis- takably to the very great importance and interest of the results that are certain to attend further research. The ice barrier dis- covered by Sir James Ross is known to be the source of the immense ice islands of the Southern Polar Sea. But it has only been seen for a distance of 300 miles. It requires far more complete examination before any approach to an adequate knowledge can be obtained respecting the extent and nature of the supposed ice cap in its rear. We know that the Southern Continent is a region of actual volcanic activity, but the extent, nature, and effects of that activity remain to be ascertained. On the Antarctic Circle land has been sighted at numerous points, but it is unknown whether what has been seen indicates small islands or a continuous coast line. Dr. Murray has pointed out that the whole Southern Continent is certainly not bounded by such an ice wall as was seen by Sir James Ross, and is not covered by an ice cap. But the extent of the ice cap and of the uncovered land is unknown. We are ignorant of the distribution of land and sea, and of ice and water in summer, and of the causes which influence such distribution. The investigation of each one of these points, and of many others, will lead to further discoveries as yet undreamt of, which must needs be of the deepest interest to geographers. There are eminent scientific men present who will no doubt refer to the results of : exploration in other branches of science. Combined together they make the exploration of the Antarctic regions the greatest and most important work that remains to be achieved by this generation. The Scientific Advantages of an Antarctic Expedition. 443 Remarks by Dr. Alexander Buchan. Dr. Alexander Buchan stated that his remarks would have exclusive reference to the first two paragraphs of Dr. Murray’s address, under the heading of “The Atmosphere,’ but more immediately to the relation between mean atmospheric pressure and prevailing winds. He supposed he had been asked to speak on this occasion from the extensive and minute knowledge of the subject he had necessarily acquired in the preparation of the reports on atmospheric and oceanic circulation which were published as two of the reports of the scientific results of the voyage of H.M.S. ‘‘ Challenger.” The former of these reports on atmospheric circulation, is accom- panied by twenty-six maps showing, by isobars for each month and the year, the mean pressure of the atmosphere and by arrows the prevailing winds of the globe, on hypsobathymetric maps, or maps showing by shadings the height of the land and the depth of the sea ; first, on Gall’s projection, and second, on north circumpolar maps on equal surface projection. The isobars are drawn from mean pres- sures calculated for 1366 places, and the winds from even a larger number of places distributed as well as possible over the globe. It may also be noted that the figures showing the averages of pressure and prevailing winds are published with the report, accompani- ments to maps of mean atmospkeric pressure and prevailing winds of the globe not yet attempted by any other writer who has spas lished such maps. This, then, is the work undertaken and published in these reports, which occupied seven years in preparing as time could be spared from official duties. The result of the charting of the pressure and prevailing winds is this: Stand with your back to the wind, then the centre of lowest pressure that causes the wind is to the left in the northern hemisphere and to the right hand in the southern hemi- sphere, a relation well known as Buys Ballot’s law. In charting the 1366 pressures and the relative prevailing winds no exception was found in any of the two hemispheres. This is one of the broadest generalisations science can point te, so far as regards what takes | place in the free atmosphere. Some years ago a theory of atmospheric circulation was published by the late Professor Ferrel, whicii, as it is not accordant with the broad results arrived at in the report on atmospheric circulation in the “‘ Challenger”’ reports, calls for serious consideration on account of its bearing on any attempt proposed to be undertaken for the ex- ploration of the Antarctic regions. One of the more recent expositors of this Nearly is Professor Davis, of Harvard College, who, in his ‘Elementary Meteorology,’ Boston, U.S.A., 1894, gives an admirable exposition of the results 444 Dr A. Buchan. now arrived at by the various workers in meteorology, and of the opinions and theories promulgated by different meteorologists in different departments of the science. The book is largely used in secondary schools and colleges of the United States, and the views there stated are widely held in America and are spreading into other countries. The following extracts from Davis’s book fairly represent these views :— ‘‘The surface winds of the temperate latitudes, and the high level currents above them, sidling swiftly along on their steep poleward gradients, must all be considered together. They combine to form a vast aerial vortex or eddy around the pole. In the northern hemi- sphere this great eddy is much interrupted by continental high pressure in winter or low pressure in summer, and by obstruction from mountain ranges, as well as by irregular disturbances of the general circalation in the form of storms”’ (p. 110). Now the facts of observation do not support the theory of the existence at any season of the year, of a low barometric pressure, or an eddy of winds, round or in the neighbouring regions of the North Pole. Observations show us no prevailing winds blowing home- wards to the region of the North Pole at any time of the year. No low barometric pressure occupies the immediate polar region in any month; but instead the opposite holds good for the four months from April to July. In April and May the mean atmospheric pres- sure is higher in the region of the Pole than it is anywhere in the northern hemisphere north of 43° lat. N.; and in June and July also higher than it is anywhere north of 55° lat. N. Now the higher pressure in these four months necessitates the existence of upper currents in order to maintain this high pressure about the North Pole. These upper currents towards the pole are exactly opposed to the requirements of the theory that the upper currents in the region of the Pole must necessarily blow not towards but from the Pole. The actual centre, in this hemisphere, north of the tropic towards which the winds on or near the surface of the earth blow, is not the North Pole; but in the winter months the low barometric de- pressions in the north of the Atlantic and Pacific respectively, and in the summer months the low barometric depressions in the Hura- sian and North American continents; and the sources out of which. the prevailing winds blow in the winter months, the high-pressure regions in Siberia and North America; and in the summer months the high-pressure regions lying northward of these continents, which, as already explained, are virtually the polar region itself. These are the facts in all regions where the winds, according to the theory, become winds blowing over the earth’s surface. The Scientific Advantages of an Antarctic Expedition. 445 As regards the southern hemisphere, Professor Davis states that : “Tn the southern hemisphere the circumpolar eddy is much more symmetrically developed.” Again, “the high pressure that should result from the low polar temperatures is therefore reversed into low pressure by the excessive equatorward centrifugal force of the great circumpolar whirl; and the air thus held away from the polar regions is seen in the tropical belts of high pressure” (pp. 110, 111)? The meaning of this is that the remarkable low-pressure region of the southern hemisphere is continued southward to the South Pole itself, the pressure diminishing all the way; and that in the region of the South Pole the air currents poured thitherwards along the surface of the earth ascend and thence proceed northwards as upper currents of such enormous intensity and volume that they pile up in the tropical region of the southern hemisphere a mean sea-level atmospheric pressure about an inch and a half higher than the sea-level pressure near the South Pole whence it has started. Now, to bring the matter to the business which this meeting of the Royal Society has in hand—if this theory were true and sup- ported by the facts of observation—it is plain that no meteorologist could signify his approval of any scheme that could be proposed for exploring the Antarctic regions, it being obvious that these strong west-north-westerly winds, if they blow vortically round and in upon the pole, heavily laden, as they necessarily would be, with the aqueous vapour they have licked up from the Southern Ocean, would over- spread Antarctica with a climate of all but continuous rains, sleet, and snow which no explorer, however intrepid and enthusiastic, could possibly face. _ But is this the state of things? Let it be at once conveded that, as far south as about 55° lat. S., the prevailing winds and the steadily diminishing mean pressures on advancing southward, fairly well support the theory. South of this, however, southerly and south-easterly winds begin to increase in frequency, until from 60° lat. S. into higher latitudes, they become the prevailing winds. This is abundantly shown from the winds charted on the maps of the “ Challenger ” Report, as well as from the unanimous experience. of all that have navigated this region from Ross to the present time. » Thus the poleward blowing winds from west-north-west in these summer months stop short, distant at least 30° of latitude from the South Pole. | These prevailing south-easterly winds necessarily imply, as has been shown in the analogous case of the North Pole, the existence of a more or less pronounced anticyclone overspreading Antarctica; which in its turn necessarily implies the existence of upper currents from the northward, blowing towards and in upon the polar region to make good the drain caused by the surface out-blowing south- 446 Sir A. Geikie. easterly winds. It may, therefore, be concluded that both the surface winds and the upper aerial currents are diametrically opposed to the requirements of this theory. What is now urgently called for is a well-equipped Antarctic Expedition to make observations which will enable meteorologists to. settle definitely the distribution of atmospheric pressure and the prevailing winds of this great region. Were this done, the position in the Southern Ocean of the great ring of lowest pressure that en- circles the globe could be mapped out; and since it is towards such a low-pressure ring that the wind-driven surface currents of the ocean flow, a contribution would thereby be made to ocean- ography, of an importance that cannot be overestimated, particularly as regards the great question of oceanic circulation, Remarks by Sir A. Getkie. Hardly anything is yet known of the geology of the Antarctic regions. By far the most important contributions to our knowledge of the subject were made by the expedition under Sir James Ross. But as he was unable to winter with his ships in the higher latitudes, and could only bere and there with difficulty effect a landing on the coast, most of the geological information brought home by him was. gathered at a greater or less distance from the land with the aid of the telescope. Within the last few years several sealing vessels have brought home some additional scraps of intelligence which only increase the desire for fuller knowledge. As regards the land, merely its edges have here and there been seen. Whether it is one great continent, or a succession of islands and archipelagos, may possibly never be ascertained. We know that in Victoria Land it terminates in a magnificent mountain range with peaks from 10,000 to 15,000 feet high; but that elsewhere it is probably comparatively low, shedding its ice-cap in one vast sheet into the sea. The rocks that constitute the land are still practically unknown. The dredgings of the ‘“‘Challenger” Expedition brought up pieces of granite, gneiss, and other continental rocks, and detritus of these materials was observed to increase on the sea-floor southwards in the direction of the Antarctic land. More recently several sealing vessels have brovght home from the islets of Graham Land, to the south of the South Shetlands, pieces of different varieties of granite, together with some volcanic rocks and fossiliferous limestones. So far as these rocks have been studied, they do not appear to differ from similar rocks all over the globe. The granites have been found by Mr. Teall to be just such masses as might have come from any old mountain-group in Europe or America. Among the specimens sent to me by Captain Robertson, of the The Scientific Advantages of an Antarctic Expedition. 447 * Active,’ from Joinville and Dundee Islands, which form the north- eastern termination of Graham Land, there was one piece of red- dish jasper which at once attracted my attention from its resem- blance to the “radiolarian cherts’” now found to be so widely distributed among the older Paleozoic rocks, both in the Old World and in the New. On closer examination, this first impression was confirmed; and a subsequent microscopic study of thin slices of the stone by Dr. Hinde proved the undoubted presence of abundant radiolaria. The specimen was a loose pebble picked up on the beach of Joinville Island. We have no means of telling where it came from or what is its geological age. But its close resemblance to the radiolarian cherts, so persistent in the Lower Silurian formations of the United Kingdom, raises the question whether there are not present in the Antarctic regions rocks of older Paleozoic age.: It would be of the utmost interest to discover such rocks in situ, and to ascertain how far their fossils agree with those found in deposits of similar antiquity in lower latitudes; or whether, as far back as early Paleozoic time, any difference in climate had begun to show itself between the polar and other regions of the earth’s surface. Among the specimens brought home by Dr. Donald and Captain Larsen from Seymour Island in the same district are a few contain- ing some half dozen species of fossil shells, which have been named and described by Messrs. Sharman and Newton, who suggest that they point to the existence of Lower Tertiary rocks, one of the organisms resembling a form found in the older Tertiary formations of Patagonia. Large, well-developed shells of Cucullea and Cytherea undoubtedly indicate the former existence of a far milder climate in these Antarctic seas than now prevails. If a chance landing for a few hours on a bare islet could give us these interesting glimpses into the geological past of the south polar regions, what would not be gained by a more leisurely and well planned expedition ? But, perhaps, the geological domain that would be most sure to gain largely from such exploration would be that which embraces the wide and fascinating field of volcanic action. In the splendid harvest of results brought home by Sir James Ross, one of the most thrilling features was the discovery of a volcano rising amid the universal snows to a height of more than 12,000 feet, and actively discharging ‘‘flame and smoke,” while other lofty cones near it indicated that they too had once been in vigorouseruption. Ross landed on one or two islands near that coast and brought away some pieces of voicanic rocks.* | If we glance at a terrestrial globe we can readily see that the * His collections are in the British Museum, but they have never been petro- graphically studied. 448 Sir A. Geikie. volcanic ring or ‘circle of fire,” which nearly surrounds the vast basin of the Pacific Ocean, stretches southwards into New Zealand. The few observations that have been made in the scattered islands further south show that the Auckland, Campbell, and Macquarrie groups consist of, or at least include, materials of volcanic origin. Still further south, along the same general line, Mr. Borchgrevink has recently (1894-95) made known the extension of Ross’s voleanic region of Mount Hrebus northwards to Cape Adare, the northern promontory of Victoria Land. He noticed there the apparent intercalation of lava and ice, while bare snowless peaks seemed still further to point to the continued activity of the volcanic fires. Some specimens, brought by Captain Jenssen, from Possession Island were found by Mr. Teall to be highly vesicular hornblende- basalt; while one from Cape Adare was a nepheline-tephrite. This region is probably one of the most interesting volcanic tracts on the face of the globe. Yet we can hardly be said to know more of it than its mere existence. The deeply interesting problems which it suggests cannot be worked out by transitory voyagers. They must be attacked by observers stationed on the spot. Ross thought that a winter station might be established near the foot of Mount Erebus, and that the interior could easily be traversed from there to the magnetic pole. But it is not merely in Victoria Land that Antarctic voleanoes may be studied. Looking again at the globe, we observe that the American volcanic band is prolonged in a north and south line down the western side of the southern continent. That it has been continued into the chain of the South Shetlands and Graham Land is proved by the occurrence there of old sheets of basalt, rising in terraces over each other, sometimes to a height of more than 7,000 feet above the sea. These denuded lavas may be as old as those of our Western Isles, Faroe, Iceland, and Greenland. But that volcanic activity is not extinct there has recently been found by Captain Larsen, who came upon a group of small volcanoes form- ing islets along the eastern coast line of Graham Land. I¢ is tantalising to know no more about them. Another geological field where much fresh and important infor- mation might be obtained by Antarctic exploration is that of ice and ice action. Our northern hemisphere was once enveloped in snow and ice, yet although for more than half a century geologists have been studying the traces of the operations of this ice covering, they are still far from having cleared up all the difficulties of the study. The Antarctic ice-cap is the largest in the world. Its behaviour could probably be watched along many parts of its margin, and this research would doubtless afford great help in the interpretation of the glaciation of the northern hemisphere. The Scientific Advantages of an Antarctic Expedition. 449 To sum up :—Geologists would hail the organisation and despatch of an Antarctic Expedition in the confident assurance that it could not fail greatly to advance the interests of their science. Among the questions which it would help to elucidate, mention may be made of the following :— The nature of the rocks forming the land of the Antarctic regions and how far these rocks contain evidence bearing on the history of terrestrial climates. The extent to which the known fossiliferous formations of our globe can be traced towards the poles; the gaps which may occur between these formations and the light which their study may be able to throw on the evolution of terrestrial topography. The history of volcanic action in the past and the conditions under which it is continued now in the polar regions; whether in high latitudes volcanism, either in its internal magmas or superficial eruptions, manifests peculiarities not observable nearer to the equator; what is the nature of the volcanic products now ejected at the surface; whether a definite sequence can be established from the eruptions of still active volcanoes back into those of earlier geological periods; and whether among the older sheets leaf-beds or other intercalations may be traceable, indicating the prolongation of a well developed terrestrial flora towards the South Pole. The influence of the Antarctic climate upon the rocks exposed to its action ; the effects of contact with ice and snow upon streams of lava; the result of the seaward creep of the ice-cap in regard to any lava sheets intercalated in the ice. The physics of Antarctic ice in regard to the ies of the Ice Age in Northern Europe and America. Remarks by Professor D'Arcy W. Thompson. The exploration of the Antarctic gives promise of gains to zoological knowledge that are in no degree less, in my opinion, than in the case of the physical sciences. The shore-fauna of the Antarctic we know only by a few scanty collections made upon the islands, especially upon Kerguelen Island; and the fauna of the deep sea is only represented by the produce of eight hauls of the “ Chal- lenger’s”’ dredge. These few dredgings gave evidence of peculiarly abundant life, indeed they were said, by common consent of the naturalists of the “Challenger,” to be the richest dredgings in all her voyage: and they were as remarkable for the diversity as for the abundance of the animals they procured. We earnestly desire, and the progress of zoological science needs, further exploration of the deep-sea fauna in all the oceans. Our knowledge of the fauna of the deep sea is only begun; it is known muchas the fauna of the 450 The Scientific Advantages of an Antarctic Expedition. shore was known a hundred years ago, and we want to know more of the undiscovered forms that are peculiar to it, and more of the struc- ture and affinities of those already discovered. But apart from innumerable facts appertaining to systematic or morphological zoology, such as every ocean has yet to yield to us, there are certain great problems of geographical distribution to which the Antarctic is peculiarly likely to give a clue. lying open to, as it were at the confluence of, all the great oceans, its fauna may co-ordinate and explain many things in the divergent faunas of the Indian Ocean, the Atlantic, and the Pacific. One such problem Dr. Murray has touched upon in his hypothesis of ‘“‘ bipolar distribution,” that is to say, of a general similarity and in many cases of actual identity between the animals of the Arctic and Antarctic Seas. This theory has been proclaimed before by others, by Théel, for instance, and by Pfeffer; but others have contested and denied it; for instance, Ortmann in regard to the crustacea, and Chun in regard to pelagic organisms. We could not have a better illustration of the poverty of our knowledge than the circumstance that so broad and clear and simple an issue as the existence or non-existence of a close relation be- tween the Arctic and Antarctic faunas should still be open to dispute. For my part I think that the bipolar hypothesis is not proven, and I am inclined to think it is untrue. I believe with Ortmann that in the Decapod crustacea at least one form is common to the Arctic and Antarctic Seas; and with Chun and others that there is evidence that similar pelagic forms occurring in the north and south, though not in the surface waters of the tropical seas, are in these latter continued across the torrid zone in the deeper and cooler levels of the sea. Ido not think that any single fish, or any Decapod or Isopod, or any certain one out of a large fauna of Amphipods known from the Antarctic Ocean, is also known from the Arctic and adjacent seas. It seems to me, however, that we have some good evidence of very curious similarities between the marine fauna of the Antarctic and that of the N. Pacific in the neighbourhood of Japan; and it may be that this is to be in part explained by the existence of a line of communication along the Western American coast, in waters singularly cold for the latitude under which they lie. We know how, in this way, such conspicuous forms as the genus Serolis, the Penguins, the Sea-elephant, the Sea-lions, and the Fur-seals, I might add the giant Sea-weed Macrocystis seem to creep up the. American shore, from what was probably their Antarctic home, to Chili, to the Galapagos, or even to North Pacific and to Japan. But these are illustrations merely of the zoological problems that Ant- arctic exploration may solve. If the bipolar hypothesis be broken down, it will only give place to other hypotheses as interesting as itself. New facts will give rise to new hypotheses, for further facts to verify Connection of Lake Tanganyika with the Sea. A51 or to disprove, and the Antarctic holds for us innnmerable problems of which we can foresee neither statement nor solution, as well as the solution of those that we can already in some measure foresee. “On the Zoological Evidence for the Connection of Lake Tanganyika with the Sea.” By J. E. 8. Moors, A.R.CS. Communicated by Professor LANKESTER, F.R.S. Received January 12,—Read January 27, 1898. (From the Huxley Research Laboratory, Royal College of Science, London.) Before 1896, when I had the opportunity of studying the fauna of Lake Tanganyika on the spot, it was known that there existed in the so-called Sea of Ujiji, one animal, the affinities of which are undoubt- edly marine. This was the medusa Limnocnida, which Dr. Boehm saw as he crossed the lake in 1883. It was known further that the jelly-fish was associated in Tangan- yika with a number of strange molluscan forms, for the empty shells of what appeared to be some six entirely new genera of gaster- opods, had been brought home by Captain Speke, Joseph Thomson, and Mr. Hore. As the animals contained in these shells have not hitherto been known, their classification by the conchologists with existing fresh-water types has always appeared extremely doubtful, and from the first Mr. Edgar Smith, who described the greater number of these forms, has held the opinion that they might eventu- ally turn out to have the same oceanic characters as the jelly-fish. it was therefore one of the objects of my recent expedition to obtain material for the complete determination of these molluscous types, and especially to ascertain if there were any other marine organisms in the lake. The results of this attempt have been to show :— I. That to the six genera of quasi-marine gasteropods, the shells of which were known, viz., T'yphobia, Nassopsis, Limnotrochus, Syrno- lopsis, the so-called Lithoglyphus, and Paramelania, there are now to be added at least two, entirely new generic forms, for which I propose the names* Bathanalia and Bythoceras (figs. land 2). We have therefore now representing the quasi-marine molluscs of Tan- ganyika eight genera of gasteropods, and to these should probably be added among the Lamellibranchiata the so-called Unio Burtoni and one of the Tanganyika Spathas.t * Diagnoses of these new genera will be found in papers now in Be hands of the Editor of the ‘ Quart. Journ. Micr. Sci.’ “+ Complete accounts of the anatomy of all these Halolimnic genera will shortly appear in the ‘ Quart. Journ, Mier. Sci.’ 453 Mr. J.E.S. Moore. Zoological Evidence for the II. That among the invertebrate population of Tanganyika there are a large number of widely separated animal types, all of which Fi¢. 1.—Shell of Bythoceras iridescens obtained living near Sumbu, Tanganyika, at a depth of 680 ft. Fie. 2.—Shell of Bathanalia Howesii, obtained near Mleroes, Tanganyika, at a depth of 950 ft. possess the same quasi-marine affinities. Thus I found that among the Crustacea there are two forms of prawns and one deep-water crab. Among the Hydrozoa the already known medusa Limnocnida. Among Porifera one deep-water sponge; and lastly there are several forms of Peridinea and Condylostoma among the Protozoa. A large proportion of these organisms are exceedingly peculiar; but others such as the two prawns, the deep-water crabs and sponge, and pos- Connection of Lake Tanganyika with the Sea. 458 sibly the pelagic Protozoa, are much more nearly related to the similar marine organisms which have repeatedly contaminated the’ fresh waters of the world elsewhere. Jt should, however, be clearly understood that even these apparently more normal types have not been found in Nyassa, Shirwa, or Kela, nor kave they been recorded from any of the Great Lakes further north. Therefore, although they are less peculiar than their associates in Tanganyika, they probably belong to the same quasi-marine, or what I shall in future call the Halolimnic group. The results of a systematic survey which was made of the geographical and bathymetric distribution of the aquatic molluscs throughout the wide area over which the expedition had to pass have demonstrated in the most conclusive manner the complete duality of the Tanganyika fauna as a whole.* In Nyassa, Shirwa, Kela, and several minor lakes, taken together, all of which I visited and dredged, there have been found the following molluscous types :— Unio, Spatha, Corbicula, Iridina, LInmnea, Isodora, Physopsis, Planorbis, Ancylus, Ampullaria, Lanistes, Vivipara, Cleopatra, Bithynia, and Melania. Not all of these fifteen genera which are now found living in Nyassa are present in the smaller lakes. In the Shirwa they are reduced to five, and in Kela they are only three. The full Nyassan series has, however, been recorded from the Victoria Nyanza, and in this more northern group of lakes there is again seen the curious reduction in the number of genera as we pass from the greater lakes to the less. From these facts of distribution it is apparent that the genera of molluscs, which occur inthe African fresh waters, are very constant over an immense area of ground. There can indeed be little doubt that the genera found in Nyassa characterise and constitute the type of the truly African fresh-water fauna as a whole. The fauna of Tanganyika appears therefore to form a striking contrast to this rule of uniformity in type which characterises the fauna of all the other lakes. Such divergence is, however, in one sense more illusory than real. All the Nyassa or Victoria Nyanza genera are found living in Tanganyika, and the fauna of this lake does not differ from the faunas of the others in kind or as a whole. It differs from them merely in there being here added to the normal series a number of molluscs which are not found elsewhere. To this superadded list, however, there attaches a unique interest, as it is entirely composed of those ten genera of gasteropods and lamelli- branchs which were instanced as possessing the same marine appear- ance as the jelly-fish and prawns. . The strange geographical isolation of the halabiteasc molluses which * The full details of my observations on the distribution of the Halolimni¢e molluscs are now in the hands of the Hditor of the ‘ Quart. Journ. Mier. Sci.’ 454. Mr. J. E. 8. Moore. Zoological Evidence for the these facts disclose is also true and characteristic of all the other halolimnic animals I have named. Itis thus rendered evident that the entire halolimnic fauna as it exists in Tanganyika now is some- thing completely distinct from and superadded to the normal African lake fauna as a whole. This fact is of the utmost import when we attempt to ascertain from what source the halolimnic animals have sprung. The isolation of these animals shows conclusively that they cannot have arisen, so to speak, de novo in Tanganyika through the effect of the conditions under which they live, for if this were so there would have arisen similar halolimnic animals under the apparently similar conditions which exist elsewhere. For the same reason they cannot be regarded as the surviving representatives of an older fresh-water stock, since were this the case we should have to believe that this old stock had been destroyed in every African lake but one. Nyassa, moreover, appears to have been a fresh-water lake longer than Tan- ganyika, yet in the Post-pleistocene deposits which occur along its shores no halolimnic fossils have been found. Now it is perhaps conceivable that prawns, which are active vigorous organisms, could by great exertions have made their way up the numerous falls along the single effluent of Tanganyika from the sea in recent times, for they have certainly thus entered many lakes already known. But with respect to the remaining halolimnic organisms, there is a singular feature common to them all which effectually precludes any possibility of this. All these animals are incapable of being directly associated with any lzving oceanic species. This fact alone demonstrates conclusively that the halolimnie fauna, wherever it came from, must be old. It has either had time to modify into its present condition from forms which are already known, or, what is more probable, it has more or less adhered to the characters of the older types from which it sprang. Delicate organisms, such as the Medusa, could not have found their way up the effluent as it now exists: it is barely conceivable that they can have been carried ~ overland, while it is altogether out of the question to suppose that either of these processes could account for the presence of the halo- limnic molluscs in the lake, as these are almost exclusively deep water forms. The genus Typhobia and the genus Bathanalia are generally beyond the hundred fathom line. Limnotrochus and Syrnolopsis are never found in less than 200 feet, and they occur up to 700 feet. The morphology of these molluscs is therefore of the first importance in determining the nature of the halolimnic group, for if the affinities of these organisms have been misinterpreted, and if in reality it can be shown that they have been derived from ancient oceanic types, they must have made their way into Tanganyika from the sea under widely different conditions from those which now exist; in fact, the Connection of Lake Tanganyika with the Sea. 455 proof of their oceanic character will more or less necessitate the idea that the Tanganyika region of to-day must have approximated in character to an arm of the deep and open sea in ancient times. During my late expedition I was able to obtain sufficient material for the complete morphological investigation of all the halolimnic molluses, the shells only of which have hitherto been known, as well as for the two new genera Bathanalia and Bathoceras represented in figs. land 2, and as I have worked over in detail a considerable number of these forms I am now in a position to state definitely what they really are. In 1857 S. P. Woodward, when describing the shells of the so-called Lithoglyphus of Tanganyika, which had been obtained by Speke, observed ‘“‘the univalve . . . . so much resembles a Nerita or Calypirea that it would be taken for a sea shell if its history were not so well authenticated,’ and similar reflections were made by other observers when describing the shells more recently obtained by Captain Hore. But possibly owing to the weight then attached to Murchison’s geological speculations respecting the African interior, undoubtedly to the fact that the ‘Tanganyika jelly-fish was not then known, and also because the fresh-water habitat of these molluscs was indisput- able, the idea of their marine origin which was thus distinctly before the minds of older zoologists subsequently became entirely obscured. The shell of Typhobia was hesitatingly classed by Smith in 1881, and more definitely by Fischer in 1887, with the Melanidz* as a subsection of that group. The shells of the Para- melanias were regarded as nearly related to the same, while the really unique so-called Lithoglyphus of Tanganyika was equally misplaced. The mere fact of the jelly-fish being, as I ascertained, associated with other marine organisms in Tanganyika would throw suspicion on these purely conchological determinations, and the actual anatomical character of the halolimnic molluscs entirely con- firms this view. The Typhobias are utterly unlike any Melanza the anatomy of which is known. These gasteropods in the character of their radule and their alimentary canals, in the presence of a crys- talline style and an anterior stomachic ccecum, in the possession of a well-developed posterior and anterior syphon, in the form of the gills and osphradium, in the position of the anal, genital, and renal apertures, as well as in the gross details of their reproductive apparatus, most closely approximate to Strombus and Pteroceras. The same inference may be gathered from the longi-commissurate character of the nervous system, while in the absence of aright pallial anastomosis, as well as in the form of the subintestinal ganglionic * In 1881 Smith became acquainted with the operculum of Typhobia, which seemed to confirm this opinion, but it is evident he doubted its correctness from statements on the same page. (‘ Zool. Soc. Proc.,’ 1881, p. 298.) VOL. LXII. 21 456 Mr. J. E.S. Moore. Zoological Evidence for the trunk, the Typhobias undoubtedly approximate to the Solaride and possibly to the Scalarids. In fact, the structural tout ensemble of the . Typhobias leaves little room for question that these gasteropods must be regarded as forms closely similar to a Pteroceras with a non- specialised foot. What is true of the Typhobias is also true of the allied genus Bathanalia, except that this form is in some respects more primi- tive, and is certainly less specialised in its shell. The so-called Lithoglyphus zonatus, L. neritinoides, and L. rufofilosus are seen at once, when anatomically examined, to have been perhaps even more completely misplaced than the Typhobias. In the characters of their radule and alimentary canals they approximate to the Planaaxide,* while in the possession of an anterior stomachic coecum and style, they show undoubted affinity to some members of the Strombide. In the character of their nervous system they are undoubtedly akin to the marine Zygoneurous Cerithia on the one hand and the longi-commis- surate Struthiolariide on the other. But the most remarkable ana- tomical feature which these forms possess is the existence in the female of an enormous epidermal invagination of the body wall beneath the eye (fig. 3), into which the embryos descend from the female genital aperture along a deep groove, and I have now complete evidence for regarding this groove, which is present in both sexes as truly homologous with the similar genital grooves possessed by the Opisthobranchs. The affinities of the new genera Bythoceras, Paramelania, and Nassopsis, are much more difficult to determine, but there is no doubt that in the curious condition of their nerves and in the general features of their anatomy they are extremely primitive. The whole nervous system of these forms, in the forwardly elongated character of the pedal ganglia and in the relation and characters of the cerebral and pleural ganglia and their connectives, actually approximate to that of a Cyclophorus. In other respects it resembles that of Trzion. Lastly, the one entire Limnotrochus which I possess seems to be nearly akin to the Paramelanian group, but the anatomy of this form will require more fully working out by sections than has yet been done. Thus, although I am not yet able to give a complete statement of the character of all the halolimnic molluscs known, enough anato- mical work has now been done for this preliminary communication to indicate clearly what will be the entire result of the investigation. It has been seen that the theory of the marine origin of the * Tt is remarkable that representatives of this family abound in the Indian Ocean and on the East African coast, the so-called Lithoglyphus of Tanganyika affording one among many instances of similarity between the molluscan fauna of hehe ika and that of the Indian Ocean. Connection of Lake Tanganyika with the Sea. 457 Fie. 3.—Semidiagrammatic representation of the reproductive apparatus in the female of the so-called Lithoglyphus rufofilosus. OV, Female genital gland. FA, Opening of oviduct. GR, Genital groove. K, Opening of brood pouch. S, Ova contained in brood pouch. i, Gsophagus. halolimnic fauna of Lake Tanganyika is entirely supported both by the facts of distribution and by those of the morphology of the individual halolimnic forms. like the medusa, the halolimnic gasteropods combine the characters of several modern marine types, aud so they cannot by any possibility be regarded as the forerunners of the modern fresh-water stocks.* Consequently, the only way in which their existence in Tanganyika can be accounted for is through the supposition that this region was, as Thomson supposed, at some tinie in open connexion either on the east, west, or north, with a deep arm of the sea. Such a conception is, however, in the most uncompromising conflict with the views respecting the permanence of the African terrestrial conditions which were advanced by Sir Roderic Murchison in 1852,7 and which have been more recently and so ably advocated by Dr. Gregory in 1896.t Nevertheless, the theory of the marine * It is certain from their anatomical characters that some of the halolimnic. molluses (the Typhobias) originatei from marine ancestors later than Cretaceous times, for they possess the characters of genera such as Strombus and Pteroceras, 1.€., genera that are Post-cretaceous and marine. Of the latter of these genera M. Fischer indeed remarks: “L/’existence de ce genre & l'état fossile parait douteuse” (‘Manuel de Conchyliologie et Paléontologie Conchyliologique,’ p- 671). + ‘Roy. Geogr. Soc. Journ.,’ vol. 24, 1864, pp. clxxv—clxxviii. ~ In his work, ‘ The Great Rift Valley,’ p. 214, Gregory restates the geological position as follows :—“ That part of Murchison’s theory, which affirms that Central Africa has never been below the level of the sea, is still in harmony with the known facts, for no deposits of marine origin have as yet been found in the interior.” ER ODT Ten eR LS Pee a g- (458 Connection of Lake Tanganyika inith t mass of the strongest kind of zoological evidence it is possible to obtain, while the above geological speculations to which it is diametrically opposed are based at best merely on the continued absence of all definite information respecting the past geological history of the ‘“ far interior” of any sort or kind. OBITUARY NOTICES OF FELLOWS DECEASED. BrnsaMin AprHorp GOULD was born at Boston, on September 27, 1524. Heentered the Boston Latin School in 1836, and graduated from Harvard College in 1844. In 1845 Gould visited Europe to make himself acquainted with the instrumental equipments of the principal Observatories. With this object he spent about three months at the Royal Observatory, Greenwich; four months at the Paris Observatory ; a year at the Berlin Observatory ; four mouths at Altona; and a month at Gotha. He was thus brought into con- tact with most of the leading astronomers of the time, and made many lasting friendships. Dr. Gould returned to America in 1848. In 1852 he joined the staff of the Coast Survey, and was employed as ‘“ Assistant in Charge” on many important determinations of longitude. The Reports of the Superintendent of the United States Coast Survey, 1852—66, contain many valuable papers by Dr. Gould on points connected with longitude determinations. In 1855 arrangements were made between the Trustees of the Dudley Observatory and the authorities of the U.S. Coast Survey, which it was thought would prove mutually advantageous, and Professors Bache, Henry, Pierce, and Dr. Gould were appointed members of the Scientific Council to superintend the work of the Observatory. In 1855, Dr. Gould visited Europe and obtained .a new Transit Circle from Messrs. Pistor and Martins, of Berlin, and some other instru- ments for the Observatory. A new Heliometer was also ordered from Mr. Spencer, of Canastota, New York, but this instrument was never completed. On December 19, 1857, Dr. Gould was appointed director of the Dudley Observatory, but his relations with the trustees were not satisfactory, and in June, 1859, his connexion with the Observatory ceased. In 1866, Gould took charge of the Valencia end of the telegraphic determination of the differ- ence in longitude .between that station and Newfoundland; and the Astronomer Royal, Airy, afforded facilities for the connexion of Gould’s station with Greenwich. About the same time, Gould reduced the observations made by D’Agelet, at Paris, 1783— 85, with a Bird quadrant. The reductions were necessarily ditfer- ential, but they afforded positions of about 2907 stars, some of which had not been observed by Bradley, and which were valuable VOL. LXIL. b il for the determinations of proper motions. In 1866 Dr. Gould also became deeply interested in the applications of photography to astronomy. He measured and deduced the right ascensions and declinations of about 50 stars from some photographs of the Pleiades taken by Rutherfurd, and showed that the results thus obtained agreed closely with Bessel’s Heliometer measures. In 1870, on the eve of his departure for Cordoba, he again took up this question and obtaine | the relative positions of the stars on Rutherfurd’s plates of the cluster Presepe. Dr. Gould was so encouraged by his success that he made arrangements for the regular application of photo- graphy at the Cordoba Observatory. This work was, however, seriously interfered with by the breaking of his photographic object- glass in transit, and secondly, when a new object-glass had been secured, by personal difficulties and the pressure of other work; but some valuable results were obtained. The establishment of the Observatory at Cordoba, and the work executed there by Dr. Gould and his assistants, must be regarded as the most mmportant astronomical work of his life. These works in- clude the ‘Uranometria Argentina,’ the ‘Zone Catalogue,’ and the ‘ Argentine General Catalogue,’ for the epoch 1875. The ‘ Urano- metria”’ gives ‘“‘the brightness and position of every fixed star, down to the seventh magnitude within 100° of the South Poie,”’ with an atlas consisting of fourteen maps exhibiting on a stereo- graphic projection the position of the stars to the seventh magni- tude. The magnitudes are based fundamentally on Argelander’s scale. The ‘ Uranometria’ was publishedin1879. The positions of the stars in the General Catalogue are generally fixed by several observations and are accurate results; this Catalogue was published in 1886. The volumes of the Zone observations were passed through the press, after Dr. Gould had left Cordoba, by his successor, and the last volume appeared only a short time before Dr. Gould’s death. The value of the work which Dr. Gould was enabled, by his own energy and the devotion of his assistants, to carry out whilst resident in Cordoba has received the fullest recognition of his contemporaries and has placed him in the first rank of practical astronomers, But besides work of this class, Dr. Gould established, in 1849, the ‘ Astronomical Journal,’ of which he continued the editor till 1861, when its publication was suspended by the war; but after his return from Cordoba this Journal was re-established, and is at present con- tinued by his friends Dr. Chandler and Professors Asaph Hall and Lewis Boss. Dr. Gould married in 1861, Mary Apthorp Quincy, daughter of the Hon. Josiah Quincy; Mrs. Gould died in 1888, and to her memory the ‘Zone Catalogue of Stars’ is dedicated. Dr. Gould’s he stil se ee Mad beet ee ee ee ee x - i ill death, which occurred November 26, 1896, followed an accidental : fall down some steps. | There is a list of 82 papers by Dr. Gould, in the ‘ Catalogue of Scientific Papers,’ published by the Royal Society. Dr. Gould was a Foreign Member of the Royal Society, a Foreign Member of the Royal Astronomical Society, a Correspond- ing Member of the French Academy of Sciences, and a member of very many other learned societies. In 1883 he received the Gold Medal of the Royal Astronomical Society. E25: Epwarp Batiarp, who died on the 19th January, 1897, will be identified always with the Central Public Health Authority of this country, the scientific repute of which his labours have done so much toenhance. As a sanitarian, however, Ballard’s own reputa- tion was assured ere he joined, in 1871, the Medical Staff of the Privy Council Office. At that date he had for sixteen years held office as Medical Officer of Health for Islington, exhibiting through- out abilities, investigatory and administrative, which had served to render him conspicuous among the representatives of English sani- tary knowledge and practice. In Ballard, Simon, then Medical Adviser to the Government, was not slow to recognise a man of the type essential for his purpose in building up the future Medical Department of the State. It was Simon’s business, no doubt, at that juncture, to discover, as inspectors, men capable of developing into the Ballards, Buchanans, and Netten-Radcliffes of after-days. Nevertheless that he did discover these men, and as well secured their services, must redound ever to his credit in the annals of English sanitary progress. Ballard was born in 1820 at Islington, a parish with which throughout his seventy-six years of life he remained closely asso- ciated. Here, at Islington School, he received his early education ; here he was apprenticed, his “ master” being parochial surgeon and workhouse medical officer; here, for many years, he practised as a _ physician, both before and after appointment as District Medical Officer of Health ; and here also he resided during the whole of his twenty years of Government service. At the early age of seventeen, Ballard commenced, as apprentice, his medical career; at nineteen he entered the Medical School of University College. Here he seems to have at once exhibited an industry in the accumulation of facts and an aptitude for exact obser- vation of altogether exceptional order, qualities of mind which then, as in after life, rendered his work always so entirely trustworthy. His career at University College and at the University of London was a distinguished one. He graduated M.B. in 1845, and M.D. in lv -1844, winning at each of his several examinations honours in the way of scholarships, exhibitions, and gold medals. He became, as a consequence, Medical Tutor to, and later on was elected Fellow of, University College. For ten or a dozen years subsequent to graduation, Ballard prac- tised as a physician, though at the same time exercising tutorial functions and writing on professional subjects. At this stage of his career he joined the Medico-Chirurgical Society, serving the Society as Referee of Papers, as Councillor, and finally as Vice-President. But it was not until 1856, when he accepted the post of Medical Officer of Health for Islington, that he entered on what proved to be his life-work. In those days the sanitary functions alike of health officer and of local authority were ill-defined or not defined at all. Ballard had to educate himself, and to educate also his masters, in matters pertaining to public health; and this, on his appointment, he at once proceeded to do, with that singleness of heart and ignoring of self-advantage which throughout his public life distinguished the man. Thus, while inculcating on his authority the broad and general lines on which they should proceed in remedying the sanitary short- comings of their district, Ballard was hour by hour studying in infinite detail series on series of facts, local and ether, the right apprehension of which was, as he was very sure, necessary to fit him as a trustworthy and far-sighted adviser in the interests of the health of his district. It was in this way that Ballard made com- mencement at Islington of those studies of his respecting the influence of various trades on health, and respecting the relation of * Sickness ” to Mortality, which subsequently, when in Government service, he extended to the whole country, with such credit to him- self and to the Department of State which he served. In further illustration of his foresight as to questions of public health likely to arise in the future, may be cited his essay ‘On Vacci- nation: its Value and alleged Dangers.’ Therein he elaborately dis- cussed, with prescience almost, the very considerations which have only lately been occupying, for a series of years, the attention of a Royal Commission; and he delivered, in 1868, practically the same judgment on the subject as that announced in 1896 by the Royal Commission in question. Ballard’s concern for problems of the future did not, however, render him in any degree insensible to the day-by-day demands of the present, so far as his district was concerned. Current events, in their etiological and in their administrative aspects, obtained from him always their full share of attention; as, for instance, the now historic Islington outbreak of “ milk-enteric fever,” in 1870. At that date, notwithstanding previous observa- aie tion by Dr. Michael Taylor, of Penrith, respecting milk-conveyed infection, little was with certainty known on the subject, and Ballard. himself was (as he has told the writer) altogether sceptical as to any . real relation between the fever and the suspected milk-service. Nevertheless, he set to work in his customary fashion to collect all the facts for and against causation of the fever outbreak by means of milk; and, as a result, not only did he convince himself that in the particular instance the suspected milk had actually disseminated the poison of enteric fever, but also he formulated in this connexion such an array of circumstantial evidence as to afford to other minds pre- sumption short only of absolute proof that the particular milk had had chief concern in the outbreak. His report on the subject has necessarily served as a model to later investigators. The limits of space proper for a “ notice” such as this forbid any- thing beyond mere attempt at indicating what sort of a man Ballard was. The work that he did in afterdays under the Local Govern- ment Board needs not to be set out. He will be known hereafter, as has been well said of him by Simon, as one of the chief confirmers and extenders of the sanitury science of his age; his researches on efiuvium nuisances, on food-poisoning, on infantile diarrhoea, on epidemic pneumonia, and on a variety of matters etiological and administrative, are duly recorded in the chronicles of the Medical Officers of the Privy Council and Local Government Board. As ‘regards his work, official and post-official, in later years, it may be remembered that much of it was done in defiance of increasing bodily infirmity. Intellectually, Ballard remained ever young, and bodily infirmity could not discourage or curb his efforts to master the unknown. After retiring from office he continued, as a labour of love, his researches into the etiology of diarrhceal disease, and he left, under his will, the data thus accumulated by him to the Medical Officer of the Local Government Board. On this subject he was still working day by day when attacked by capillary bronchitis, which proved fatal in the short space of two days. Ballard was destitute of personal ambition. He was content with his work, and with appreciation of that work by those competent to judge of it. He sought neither bonours nor emolument. From Government he obtained no honours at all; and such recognition as jus work brought him from Medicine and from Science was but tardily bestowed. Wer Ee ALEXANDER Heyry Green, M.A., F.R.S., F.G.S., Professor of Geology in the University of Oxford, was born at Maidstone, October 10, 1832. He was the son of the Rev. Thomas Sheldon Green, formerly Fellow of Christ College, Cambridge, who at the VOL. LXI. ¢ V1 time of the birth of his son Henry was head master of the Grammar School, Ashby-de-la-Zouch.* 7 Green received his early education in his father’s grammar school at Ashby; and the man who originally inspired him with his life-long regard for geology was the Rev. W. H. Coleman, one of the masters in the Ashby School. Coleman and his pupil were fortunate not only in living in a district of exceptional geological interest, but also in the circumstance that a few years previously (1834) Mr. Mammat, a local geologist, had made the structure of the neigh- bourhood classic in his well-known work ‘Geological Facts.’ Coleman, who seems to have been not only a scientific enthusiast but a man of singularly loveable character, died young. His memory was fondly cherished by his pupil, who always spoke of him in after years with great personal affection, and dedicated to his memory his own work on ‘ Physical Geology.’ In the course of a few years Green made himself so familiar with the structure of the Ashby country that when Professor (after- wards Sir Andrew) Ramsay paid a visit for the purpose of examin- ing the district on behalf of the National Geological Survey, he was at once referred to Green as the acknowledged local authority on the subject. Ramsay soon convinced himself of the young man’s remarkable scientific abilities; and, always on the lock out for exceptional geological talent, snemiabed that if he desired to devote his life to geological work he should become a candidate for a post upon the National Survey. Frora Ashby School, Green proceeded to Caius College, Cambridge. In 1855 he was placed sixth among the Wranglers in the Mathema- tical Tripos, and was elected a Fellow of his College in the same year. During his tenure of his Fellowship he tock pupils in mathe- matics at his own College, and afterwards held masterships in mathematics at two private schools. But although mathematics had gained for him his high position in the University, and the teaching of mathematics had a special charm for him, yet the science of geology had obtained too strong a hold upon his mind and his sympathies to be relinquished. The love of the science which had been awakened in his early life by Coleman, and his own keen interest and pleasure in the pursuit of personal geological investigation had both become intensified and fixed at Cambridge by the eloquent teachings of Sedgwick; and in 1861 Green apphed for, and was appointed to, a post on the Geolo- gical Survey of England and Wales. His connexion with the * T have to thank Professor T. G. Bonney, F.R.S., Mr. H. B. Woodward, F.R.S., Mr. W. W. Watts, M.A., and others, for Eaeeion and kindly assistance in the preparation of this memoir. Vii Survey lasted from 1861 to 1874. He ranked as an assistant geologist from 1861 to 1867, and as a geologist to the close of his Survey career. During the time of his connexion with the Survey, he was engaged first in mapping the Jurassic rocks in the Midlands, and afterwards in surveying the Carboniferous rocks in Derbyshire, Yorkshire, and the bordering counties. The broader results of his field work became embodied in a large number of published maps of the Geological Survey, some upon the l-inch and some upon the 6-inch scale. But several detailed survey memoirs, descriptive of the country surveyed by himself and his colleagues, were written wholly or in part by him. Among these may be mentioned ‘The Geology of Banbury ” (1864), “Geological Description of the Country round Stockport” (1866), “Tadcaster” (1869), ‘‘ Dews- bury” (1871), “ Barnsley ” (1878), and “ Wakefield ” (1879). A memoir on the geology of North Derbyshire, of which the first edition was issued in 1869 and the second in 1887, was also written chiefly by Mr. Green. But his most important survey publication was his “Geology of the Yorkshire Coalfields,” issued in 1878. This work is one of the largest, and from an economic point of view, certainly the most important memoir of a single coalfield yet pub- lished by the Geological Survey of England and Wales. It gives an exhaustive account of the structure and economic aspects of all those parts of the great Yorkshire coalfield which lie around the chief manufacturing centres of that county. It is a complete record of the detailed work laboriously accomplished by Mr. Green and his colleagues on the Survey, and it contains one of the best descriptions of the British coal measui‘es in the language. From this time forward Mr. Green became naturally regarded as one of the leading British authorities upon all geological matters connected with coal and coal mining, and the relationship of geolo- gical structure to economics in general. In 1870 he resigned his post on the Survey, having been appointed Professor in Geology in the newly founded Yorkshire College at Leeds. Students of geology, however, were few in number, and funds were not superabundant; but Gveen’s energy and abilities found ample scope, for, in addition to holding the geological chair, he was appointed to the Professorship of Mathematics in the same College, and he acted also as lecturer upon the subject of Physical Geography. Professor Green’s connexion with the Yorkshire College ceased in 1888, when, upon the resignation of Professor Prestwich, he was appointed to the Chair of Geology in the University of Oxford. He threw himself with his characteristic energy heartily into his new | work, giving not only the ordinary courses of lectures and laboratory vill instruction, but holding special excursion classes and delivering occasional courses of lectures to extra-University students. Early in 1895 he suffered from a severe attack of influenza. From this he never seems to have fully recovered, and in his enfeebled state the ceaseless strain of work and responsibility soon began to tell severely upon his general health. Early in August, 1896, he had a paralytic stroke. This affected his right side and confined him to bed; but he wrote to his friends cheerfully, throagh an amanuensis, that he was making good progress, and hoped before many weeks were over to be at work again. But these hopes were, alas, illusive, and a second attack on August 19 proved fatal. Professor Green was M.A. of both Cambridge and Oxford, and a Fellow and Vice-President of the Geological Society of London. In 1886 he was elected a Fellow of the Royal Society, on the Council of which he served in 1894 and 1895. In 1890 he filled the office of President of Section C (Geology) at the meeting of the British Association at Leeds. He was Examiner in Geology to the Univer- sities of Durham and Cambridge, Examiner for the Home and Indian Civil Services, Assistant Examiner in Physiography to the Science and Art Department, and, at the time of his death, one of the Exami- ners in Geology for the University of London. Professor Green contributed occasional original papers to the Geological Society and to various scientific magazines. Among these may be mentioned his papers on the “‘ Carboniferous Rocks of the North of England” and on “ Sub-aérial Denudation,” “The Geology of Donegal” and “The Geology of the Malvern Hills.” But his publications in this department of geology, when compared with those of many of his geological contemporaries, are relatively few and insignificant. On the practical side of the science, however, there were not many who accomplished so much or whose reputation was so widespread or so well deserved. His life-long acquaintance with the details of geological structure, his clear head and his sound judgment, rendered his advice on matters of engineering geology of great value; and, as a consequence, his services were being continu- ally called into requisition in economic undertakings, especially those in reference to coal mining and to water supply. He was engaged in many of the most important operations of this nature carried out in Britain during the last thirty years. He also visited parts of South Africa in the pursuit of his practical work, and in this way he was able to obtain a considerable insight into its little- known geology. The paper in which he brought his African results before the Geological Society is one of the most important contribu- tions yet made to the subject. Professor Green wrote papers upon the Yorkshire coal measures in a volume of ‘Hssays upon Scientific Sabjects,’ issued by the 1X Professors of the Yorkshire College, and he published a popular little work upon the ‘ Birth and Growth of Worlds.’ But the book by which Green was most widely known among geologists and lovers of geology was his ‘ Manual of Physical Geology for Students and General Readers,’ the first edition of which was published in 1876, and which attained to a third edition in 1883. This manual is admittedly the best English work in this branch of the science. It is remarkably typical of its author, thoroughly practical and almost painfully conscientious; it is lucid and modest, but at the same time original and bold; and it has done more, perhaps, than any other work to foster a wide appreciation of physical geology. In this work we see Green at his best. His deep love for his science and its students is evident in every chapter. His delight in the beauties of literature and in good literary style gives a peculiar charm to the book as a whole; and the mathematical bent of his mind is evident in the crystallographic part of the work and in his original methods of representing outcrops of strata and the like, methods which have subsequently proved of especial utility in the advance of geological science. Professor Green’s geological teaching in his lectures, though not perhaps calculated to move to enthusiasm, was exact and thorough. His lectures were usually illustrated by careful experi- ments, and were directed less to the presentation of geological facts and accepted theories than to the inculcation of the true scientific method of research and habit of reasoning. But as an original worker in and as a teacher of practical field geology Green had few equals. His own life training, his special artistic abilities, and his love of detail all conspired to this end; and there can be no question that it is to his labours and teaching that British stratigraphical geology owes much of its present influence. In the death of Professor Green not only has Oxford University lost a most distinguished member of its professorate, but geology has lost a steady and fruitful worker, one who loved the science entirely for its own sake, and whose life was spent in quietly labour- ing for its advancement. His wide knowledge of hterature and of © general science, and his deep and unbiassed appreciation of all modes of their progress, made him a most pleasant companion; while, his calm judgment and his cautious habit of mind rendered him an invaluable colleague. Always frank and fearless in the expression of an opinion at which he himself had arrived, he was, nevertheless, so hearty, so genial, and so unselfish that his loss will be most keenly felt, not only by all his fellow workers in the science, but by a still wider circle of loving friends. C. a VOL. LXII. ad x EpwarD James STONE was born in London on the 28th February, 1831. As a child his constitution was delicate, so that soon after entering the City of London School his health broke down, and he was sent to the country for several years, where he was educated at a private school until ready to enter King’s College, London. Although his higher education only began at the age of 20, he took a scholarship at Queens’ College, Cambridge, in 1856, whence he graduated as Fifth Wrangler in 1859, and was immediately elected to a Fellowship. The following year he was selected for the important position of Chief Assistant at the Royal Observatory, Greenwich, his predecessor in that post, the Rev. R. Main, having been appointed Radcliffe Observer at Oxford. With what diligence he applied himself to the duties of his office and how wide a view he took of his responsibilities, is made evident by the series of important papers which he soon afterwards began to communicate to the Royal Astronomical Society. His work was obviously congenial, he had a marked inborn capacity for dealing with large masses of figures, a high estimate of the practical importance of the work on which he was engaged, sound mathematical training, and an almost impatient desire to derive from the long series of Greenwich observations results which would be of immediate value to science. It is not difficult to trace the origin of the line of research which Stone subsequently followed with such zeal and pertinacity. Fresh from study of the lunar and planetary theories, so far as these subjects were treated in his Cambridge curriculum, the attention of the young astronomer was early arrested by the interesting problems which were then opening up in consequence of the researches of Hansen and Le Verrier. We have a statement of the motif of Stone’s first work in the introductory paragraphs of his first astronomical paper,* ‘“ Deter- mination of the Solar Parallax from N.P.D. Observations of Mars at Greenwich and Williamstown.” “Tn his tables of Mars, published in the ‘ Annals of the Imperial Observa- tory,’ Paris, 1861, M. Le Verrier remarks that it is impossible to reconcile the observations of Mars with theory without attributing to the perihelion a motion greater than any which can be obtained except by a sensible increase of the received planetary masses; that the necessary agreement between theory and observation could be obtained by increasing the received value of the mass of the earth in proportion to the sun’s mass by not less than a tenth part, but that such an increase in the received value of the earth’s mass would necessitate a corresponding increase in the received value of the sun’s mean equatorial horizontal parallax of a thirtieth part. ‘© M. Le Verrier deduced the same result from a discussion of the latitudes * “Monthly Notices R.A.S.,’ vol. 23, 1863, p. 183. Xl and motion of the node of Venus. The difficulties raised in the theory of Mercury, although not removed, were slightly diminished by the same increase of the earth’s mass. “In his solar tables, M. Le Verrier has adopted the value 8°95” for the mean equatorial horizontal solar parallax, this value was obtained by deter- mining from observation the coefficient of the lunar equation, and assuming the mean lunar parallax and data furnished by the theories of precession and nutation. “The way in which M. Le Verrier has thus evolved from the theories of Venus, the Earth, and Mars, the necessity of the value of the mean solar parallax much greater than the usually received value 8°57”, and not differing greatly from 8°95”, must render it extremely probable that the true value of the sun’s mean parallax does not differ greatly from that quantity.” Stone then proceeds to discuss the Greenwich and Williamstown observations, and derives a value of 8°932" for the solar parallax. In the ‘Monthly Notices’ for May, 1865, Stone remarks that in the lunar theories of Plana, Pontécoulant, and Lubbock the co- efficient of the parallactic inequality deduced with the usually received values of the involved constants amounts to 122°1", which, if we increase the value of the mean solar parallax by a thirtieth part, becomes 126-2", coinciding closely with the observed values as derived by Airy, viz.:—124°7" from meridian and 125°5” from altazi- muth observations. He suggests that it would be a point of interest to determine whether Hansen’s lunar theory would bear any con- siderable increase in the mean solar parallax. | Stone had apparently overlooked a letter of Hansen’s,* in which the writer says :— “ The coefficient of the parallactic equation I found to be 125°705”, “an amount exceeding any which has hitherto been assigned, and which indicates a greater value of the sun’s parallax than has been deduced from the observations of the transit of Venus. The Greenwich observations, exclusive of any others, assign the foregoing value of the parallactic in- equality, and the Dorpat observations nearly the same value. I cannot, therefore, alter it.” In the following number of the ‘ Monthly Notices,’+ Hansen, in reply to Stone, refers to the above quoted paragraph, and in the same periodical for November, 1863, gives with more detail, as the result of his researches, the value 89159” for the mean solar parallax. Till the note of alarm was thus sounded by Hansen in 1854, and echoed by Le Verrier in 1861, astronomers had almost universally * “Monthly Notices R.A.S.,’ May, 1854. + June, 1863. d 2 Xi accepted as definitive the value of the solar parallax found by Encke from his discussion of the Transits of Venus of 1761 and 1769, viz. :— 8: de Winnecke, who proposed the practical programme for the observa- tions of Mars in 1862, discussed the thirteen corresponding obserya- tions made at Pulkowa and the Cape,* and found for the solar parallax 8964", This, together with Stone’s result already quoted, seemed to prove, from the practical as well as from the theoretical side, that the then accepted value of this fundamental constant of astronomy was probably at least one thirtieth part of its amount in error. The question of the exact redetermination of the solar parallax became at once one of supreme interest, and, as Airy put it, “ the noblest problem in astronomy.” Stone threw himself into its divonantat with his characteristic vigour. In the ‘Memoirs of the R.A.S.’+ he discusses afresh the meridian observations of Mars made at Greenwich in 1862, com- bining them with those made at Williamstown and the Cape of Good Hope, and derives for the solar parallax © 8945", In May, 1865, he communicated to the R.A.S. an important memoir on the “ Constant of Lunar Parallax,” in which he derived the value of that constant, as defined by Adams, from a series of observations of the moon made at the Cape Transit Circle in the years 1856-61, combined with corresponding observations made at Greenwich. Up to the present time Stone’s result is accepted by astronomers as the most reliable direct determination of this impor- tant constant. In the ‘ Monthly Notices’ for April, 1867, Stone calls attention to a slight numerical error in the computation of the value of the mass of the moon which was employed by Le Verrier in his derivation of the solar parallax from the known relations between the parallaxes of the sun and moon, the mass of the moon, and the lunar equation.{ With this revised value of the moon’s mass (computed from Le Verrier’s adopted values of the constants of precession and nutation) and with the value of the lunar equation (6°50") derived by Le Verrier from his discussion of the existing meridian observations of the sun, Stone derives for the solar parallax the value 8-917, * ¢ Ast. Nach.,’ 1409, April, 1863. + Vol. 33, May, 1864. t ‘ Annales de l’Observatoire de Paris,’ vol. 4, p. 101. x1 In two notes* Stone reverts to the determination of the moon’s mass, and, adopting 50°377" for the luni-solar precession, and 9°223” for the constant of nutation, he derives for the moon’s mass the value phew 8136" Following the same general line of research, Stone then under- takes a direct determination of the constant of nutation based on Greenwich N.P.D. observations of Polaris, 51 H Cephei, and 6 Urse Minoris, and the R.A. observations of Polaris made with the transit circle 1851-65. The preliminary results of this work are published in the ‘ Monthly Notices,’ vol. 28, 1868, p. 229, and vol. 29, 1869, p- 28. The complete discussion appears in the ‘ Memoirs of the R.A.S.,’ vol. 37; the resulting value of the constant of nutation is 97134", In the concluding portion of this memoir, Stone draws attention to the strong evidence which his discussion gives of a periodic change of latitude, a subject which, had he followed it up, might have led to an earlier discovery of Chandler’s now well-known law of variation of latitude, but his thoughts were otherwise occupied. In the supplementary number of the ‘Monthly Notices’ for 1868, Stone communicated to the Royal Astronomical Society his ‘“ Re- discussion of the Observations of the Transit of Venus of 1769,” in which, after reverting to the above-quoted evidence in favour of an increase in the accepted value of the solar parallax, he proceeds to rediscuss the observations made at five stations where internal con- tacts of Venus with the sun’s limb were observed both at ingress and egress by ten observers. An independent interpretation was put upon the language employed by each observer to describe the phenomena which he noted at different instants of time, and the assumption was made that such phenomena could be divided into two distinc! classes—viz., true and apparent contacts—separated by a definite interval of time. This interval (assumed to be constant for all observers) was introduced in symbolical form into all the equations. The solution of the equations so formed led to the value ak. for the solar parallax, with the estimated probable error 002". | The observations were there represented as follows :— * “Monthly Notices R.AS.,’ vol. 28, pp. 21 and 42. X1V Computed Computed minus minus Observer. observed. Station. mean. Helloows ‘ —i7s Seve aa Wardhus...... 0°68 Sajmovics ..... +2°8 j ae 1 Wales eeecerveee +0°4 \e d , B ; (IT Dymond yabeon —0'6 me Chappe iar Anhydride, and Nato Oxide. Mt ORD RayizicH, ¥.R.S. . “ ; , : ' 4 ‘ plication of Harmonic Analysis to the Decal Theory of se) 8. Part IT. (On the general Integration of Laplace’s Dihareteat ae By 8.8. Hove, M.A., Fellow of St. John’s College and | . Communicated — eth . Piece. By Z «apa Asnwourn, B.Sc. Commune oy ny cay ae aati fils pug : ! q SAN | in gh Wall of the Body-cavity of Vertebrates. By Bowell Se. aes A Wah College, Cambridge. Communicated by . ° ° . « ° i) aging) and a oa Coaxial Pee ae) Cylindrical Current Sheet, ‘By J. ‘Vierace as ONES, ,# ce s. : | FARD Jaane Stone i. _ PROCEEDINGS 0 “THE ROYAL SOCIETY. CONTENTS. ‘Note on some further Determinations of the Dielectric Constants of Organic _ Bodies and Electrolytes at very Low Temperatures. By James Dewar, _ MA, LLD., F.BS., Fullerian Professor jof Chemistry in the Royal + Institution, and J. A. Fremine, M.A., D.Sc., F.R.S., Professor of Elec- a As 5* “Meeting of December 16, 1897, and List of Papers read . . 2 : : On a Method of determining the Reactions at the Points of Support of Con- Fi tinuous Beams. By Gzorert Witson, M.Se., Demonstrator in Engineer- a Rs gin Communicated by Professor OSBORNE REYNOLDS, F.R.S. ‘ : ; er “The Comparative Chemistry of the Suprarenal Capsules. By B. Moorg, M.A., Association Research Scholar. Communicated by Professor SCHAFER, MP gS eae Se ke? ORT el Mauctet yas | Memoir on the Integration of Partial Betepential Equations of the Second _ does not exist in general. By A. R. Forsytu, F.R.S., Sadlerian Professor _ imtheUniversityof Cambridge - - +. + + + - “on the Biology of Stereum hirsutum, Fr. By H. MarsHatt Warp, D.Sc., F.B.S., Professor of Botany in the University of Cambridge . - ‘<4 On the Thermal Conductivities of Single and Mixed Solids and Liquids, and their Variation with Temperature. By Cuarizs H. Lens, D.Sc, Assistant Lecturer in Physics in the Owens College. Communicated by — Better OURTATER IRS on “¢ Cloudiness : Note on a Novel Case of Frequency. a Karz Pearson, M. A: a PRS. University College, London . Ag Basra Reng On the Occlusion of Hydrogen and Oxygen by Palladium. By Lupwie Mond, Ph.D., F.R.S., Witt1am Ramsay, Ph.D., LL.D., Sc.D., F.BS., inh and JoHN SHIELDS, D. eee Ay eink 42 ty BN Tortus RIESE Crh Om ott Rice Ni ad eee peeMRRe Ream grails o WT Sa ‘Price Two Shillings. a trical Engineering i in University College, Pandan \ : : : - ing in the Whitworth Laboratory of the Owens College, Manchester. — _ Sharpey Research Scholar and Assistant in Physiology, Waianae . College, London, and Swate Vincent, M.B. (Lond.), British Medical — 280 Order in Three Independent Variables, when an Intermediary Integral — 285, Cae 268 283i} OFFICE LIBRARY OR wenn A i Smithsonian uistiution, termination of the Indices of Refraction of yarious Substances for. oi ea ITI. Index of Refraction of ue a J. AGADIS a aed ‘By JAGADIS ‘nace Boss, MA. - D.Sc., Professor of “pha RAT uipomht Calcutta. Communicated by Lorp Mh CONTENTS. PAGE ¢ Vortex Motion. Part III. On Spiral or Gyrostatic Vortex egates. By W.M. Hicns, F.RS. . 2 A pie 9 a0 8 Maimeri acology of Aconitine, Diacetylaconitine, Benzaconine, and Aconine, cred i in Relation to their Chemical Constitution. By J. THxopoR# M._D., F.R.S., and Wynpuam R. Dunstan, M.A. F.RS. 2. ean Fishes. By B. Moors, M.A., Sharpey Scholar, University e, London, and Swatze Vincent, M.B. (Lond.), British Medical tion Research Scholar. Communicated by Professor E. A. ts of Extirpation of the Suprarenal Bodies of the Eel (Anguilla la). By Swate Vincent, M.B. (Lond.), British Medical Associa- search Scholar. Communicated by Professor E. A. SCHAFER, , Quadrant EHlectrometer as a Wattmeter and Voltmeter. By care tk Sat AR aoe - w: st WILSON. Communicated by Dr. J. Hopkinson, F.R.S. . ag etic Properties of almost pure Iron. By Ernest Winson. Com- 1unicated by Dr. J. Hopxinson, F.R.S. . : ; Sena: HY ow Method of determining the Vapour Pressures of Solutions. By B. H. Wave, B.A. Communicated by Professor J. J. THomson, ° SD. 2 . . . . . . ° 2 Price Two Shillings and Sixpence. * ] THE ROYAL SOCIE Fe VOU. TT. CONTENTS. PAGE Meeting of February 17, 1898, and List of Papers read . : 2 eb a ates _ Mathematical Contributions to the Theory of Evolution. On the Law of Ancestral Heredity. By Karu Prarson, M.A., F.R.S., University College, London _. Pd AR fi ote cy Ake SNe RU AVR WOR ae . 886 Mathematical Contributions to the Theory of Evolution. On the Inheritance of the Cephalic Index. By Miss Ciceny D. Fawcett, B.Sc., and Karu Prarson, M.A., F.R.S., University College, London ‘ } : . 418 Comparison of Oxygen with the Extra Lines in the Spectra of the Helium Stars, 8 Crucis, &c.; also Summary of the Spectra of Southern Stars to the 3} Magnitude and their Distribution. By Franc McCzman, F.R.S. MMC Oar cc epee re al Sh Re eee ane Ber. : Price Two Shillings. _ Marcu 16, 1898. VOL. LXII. No. 387. CONTENTS. PAGE Meeting of February 24, 1898 (Discussion Meeting) . ‘ : : « 423 The Scientific Advantages of an Antarctic Expedition. By Jonn Murray, Pees). Ph.D. ERS... ; d in is MPM Gr ii irame Pe Remarks by the DukE oF ARGYLL . j ; ; : : : . 434 Remarks by Sir J. D. Hooxer. : : : : : hs . 436 Remarks by Dr. G. NEUMAYER : : ; ; é ; ; . 488 Remarks by Sir CLEMENTS MARKHAM . : é ‘ é - 9. 442 Remarks by Dr. ALEXANDER BucHAN . : : : : : . 443 Remarks by Sir A, GHIKIE . , : : j , : i . 446 Remarks by Professor D’Arcy THOMPSON. : 4 : é . 449 On the Zoological Evidence for the Connection of Lake Tanganyika with the Sea. By J. H. S. Moorz, A.R.C.S. Communicated by Professor LANKESTER, F.R.S. . ‘ : ; ; : } i ; : Junaid Price One Shilling. Masgca 29, 1898. OFFICE Lip OR Price Sixpence. — WAL