xviii BRITISH GRAPTOLITES. 



sometimes in contact for their whole length {Gr. colonm) or for part of their length 

 {Gr. yrhnloii), or they are quite free as in Rastrifcs. 



In form, each of the cellules may be compared to " a small sac," the length of 

 which is always greater than the width, and is either rectangular or circular in 

 section. Sometimes they narrow towards their orifices, or they curve over in the 

 form of a hook {(iv. jifini^on). I^arrande considers that there is no douljt that the 

 wall between two contiguous cellules is double, and he represents this in his figures 

 of Gr. priodou. 



Each cellule, according to Barrande, has two orifices : 



(1) The mternal orifice, communicating with the common canal, and generally 

 subrectangular or round in shape ; 



(2) The external orifice, which presents many variations in form. 



He discusses the various forms of cell apertures in consideraljle detail, from the 

 simple straight apertures of Gr. bohemicns, which open upwards, through the 

 oblique and spinose ones of Gr. testis, etc., to the curved ones of Gr. p-iodon, which 

 open downwards. The apertures of several species are provided with spines, and 

 when two spines are present these are arranged either above and beloAvthe aperture 

 {Gr. cldmsera), or symmetrically on the lateral parts of the Ijorder {Gr. testis). 

 Barrande records all these spines or " ornaments " as " simple appendages of 

 the test." 



6. Nature and Ornaments of the Test. — The solid test of the Graptolites is about 

 •1 mm. in thickness, and of a horny character. Barrande believes that it contains 

 little, if any, carbonate of hme, and is inclined to think that it is largely car- 

 bonaceous ; and ho points out that the black colour of the shales in which Grapto- 

 lites occur seems to confirm this opinion. 



As regards the markings and superficial ornamentation of the test, he observes 

 oblique striations on some of the cells, but he does not appear to have realised their 

 vital importance as " growth-lines." Except in Gladiolites, the Graptolite test is 

 smooth and contiimous. With respect to the genus Gladiolites, which he was the 

 first to distinguish, Barrande recognises and describes with great accuracy the 

 peculiar network of threads forming the test, and is of the opinion that the meshes 

 were either open, or else closed by " a membrane chemically different from that of 

 Avhich the network is com})osed." 



7. State of Preser cation. — Barrande })oints out that in Bohemia tlie (]!ra})tolites 

 are usually preserved as impressions, but occasionally (in the limestones) they occur 

 in relief with the test preserved, Avhile in certain of the higher l)eds internal casts 

 alone are found. 



The accuracy of the observations of Barrande, the breadth and lucidity of his 

 conclusions and (lescri])tions, and the beauty of his illustrations, make his work not 

 (inly :i cljissic, liiil nlso most- valiiahle For reference even at the present day. But 

 liari'ande s work, impDrtunt as it was, was only a stage in the progi'ess of grapto- 



