TRAN.SACTIONS 01' SECTION I,'. 



rn.'i 



On """"S rrinauifi af Fi'sh/rovi. tlir J^/i/irr Sihiridii Rorlx if Penvf>i/loaniat 

 ' lly IVofessor" I*;. W. Ci.ayidm;, 7;..l., n.Sc. {Lo,rl.), F.G.S. 



Till' eurlifst verlcLrate iinimals yet known from any liarf of tlin world iiro 

 fiomt) ri'iuiiins of lisli in tho rpper Silurian mck.s of iuiiiliinil. Tlu-y are for llit! 

 most imrt of three tyju's. I'irst, Nliort, lin-vSj)int'M, nunicd liy A^nssiz Onc/ins 

 tciithtriiifii''^ ; st't'ond, frai;nients of sIia^'-recM, or tla' .skin of a placoid ti>h 

 if/i,'l,iilii.i and SplKKjodus), l)t'lon<j:iii)^ ]irolm))ly to tlic same tliat carried tlies])iiie; 

 and tliiid, ovate, finely striated plates or sliielils, supposed to lie the defensive 

 armour "f some tisli, unlike any now livin^'. 



Xo one has doubted the ichtiiyic nature of the iirsi and second of tlie.*e tiiroe 

 I'limis*. ''"t iw rej/anls tho third there has been nnich Cdulroversy. Kvidently 

 iilhi'il til Ci-p/uil'ixpit, its rij.'hl to the name of fish has lieen called in qucistion, ami 

 i-tisniciiiii has been raised in rej/ard to the whole family of the ( 'ephalaspids. 



On the whole, however, it seems best to retain them in the class of lishes, and 

 ti) this couolusion I'rofes.sur Jluxley evidently inclines in the conclusion of iiis 

 'Kssiiv en the Classilication of the Devonian l'"ish.' Out; may expect some, or 

 even cunsiderable, divergence of structure from the usual iclubyic tyjies in such 

 earlv furnis. 



TIk's" J'higlisb fossils occur in the lowest be<ls of the Devonian (Cornwall), and 

 in the hi"hest beds of the Silurian (Sliro])shire and Hereford). The well-known 

 rpiiir Ludlow * bone ))ed ' bus yieldeii them in considerable ((uantity, and one 

 Miirinion is rejiorted by Sir ('. ij\v\\ in his 'Elements of (ieolofry ' (Irttio) as 

 'disiMivered from the JiOwer l^udlow, beneath the Aymt ; try limestone.' lielow 

 this liori/on 1 liave never heard of (heir occmn'ence. 



The lMij.dish Jjudlow, taken as a whole, has been usiuilly correlated with the 

 Lower Helderber^ of Nortli America, and on j,'oo(l ^round.s, both containin;r 

 Eurin>tcni.-i and l'ffri/(/i>fns. 1'he ]in;.flisli Lower Ludlow and the \\'ater-Lini& 

 or basal beds of the Is'ortb Anu'rican ]x)wer Ilelderberji- are the lowest strata 

 eontainin;; these fossils. On both sides of the Atlantic they range from this level 

 upwards into the Devonian. 



The oldest vertebrate fos.sils yet announced from America are those found 

 in the t'orniferous limestone or Lowest Devonian of Ohio. I'ossibly the beds 

 at Gasiie on the (Julf of St. Ijawreiice are somewhat lower; as they have yielded 

 Vi'pliiildi/iis, which is not yet known from Ohio, and CWco.v/^-iw, of which Ohio 

 lias vielded only a single s]iecimen. Xo authenticated tiih-1'ossil has yet beeu 

 announced from the Upper Silurian_ rocks of America. 



It is true that re])orts of the discovery of such remains liaAO been publishe(l 

 at various times, but investigation has proved them all erroneous. (See 

 ' I'ala'ontolof'V of New York,' vol. ii. pp. -ili), .'ii'O, pi. Ixxi.; ' Anwrican Journal 

 of Science,' second series, vol. i. p. (ii'; ' I'alieontology of Ohio," vol. ii. p. tlCyJ.) 



During his recent work on the I'alrcontology of Terry County, Pennsylvania, 

 the author came upon some fossils which at once suggested relation.ship to the 

 Ludlow group above described. Among tliLMn were a few spines recalling Otivhux 

 tenmfriiit.us, but with som(! ditl'erences. lit' has named them (hic/ius Pcnii- 

 ■vilvanicus. With them he discovonid abundance of specimens bearing a strong 

 resemblance to Scap/iaspis, l)ut larger, and dillering in some other respects. Thes»! 

 he named Palaa-ipi.'! {R cUlptiva &w\ P. hifruncafa). 



Comparing these with »VrY(/>/(r?.v7>/.t we lind them much thinner, not exceeding 

 one-fortieth of au inch in thickness; whereas specimens of Scaphanpis in the 

 author's possession from ('ornwall are in some places mucli thicker. The striation 

 on both is equally fine, but is rather less regular on the American specimens. 

 These also show no trace of the spine in which the shield of Scaphaspis terminates., 

 as shown by Murcbison in ' Siluria.' 



No traces of the English fossil .shagreen — T/ielo'Ins and Sphifiodus — have been 

 found in the Penn.sylvanian beds, though it abounds in the Ludlow rocks. 



The fossils were found in a bed of sandstone about 200 feet below the base of 

 the Water-Lime in I'erry county, Pennsylvania ; near the top of the great ma.ss of 

 variegated shale composing the Fifth Group of Kogers in the First Survey of 



'Ii 



