64 INVERTEBRATE PALAEONTOLOGY 



destroyed in making the next in sequence. The specimen 

 should be rigidly fastened to a flat plate, and ground 

 away until signs of the required structures appear. The 

 surface should then be cleaned (preferably by a wash 

 of dilute acid), and photographed or carefully drawn. 

 Grinding may then be resumed until a measured thick- 

 ness has been removed, and the delineation repeated. 

 If the amount destroyed at each episode is recorded, 

 and the representations of each surface are to the 

 same scale, models in wax or card can be constructed. 

 Special apparatus has been devised for this process, but 

 it can be carried out with fair accuracy by hand after 

 some practice. The method is well adapted for study 

 of Brachiopod or Pelecypod hinges, although it has 

 been employed most frequently on Vertebrate skulls. 



A simple and valuable modification of grinding was 

 perfected by the Rev. Norman Glass for display of 

 Brachiopod brachidia ; the results of his process can 

 be appreciated in the illustrations to Davidson's mono- 

 graph. The tightly closed valves of Brachiopoda often 

 failed to admit fragmental debris, so that their cavity 

 became filled with crystalline calcite introduced in 

 solution. The softness and transparency of this infill- 

 ing, deposited so gently that no internal structures are 

 damaged, makes possible display of brachidia by scrap- 

 ing away the shell and surplus calcite, polishing the 

 residue (when the opaque shell-ribbons are reached) by 

 dilute acid. 



Natural moulds of Conchifera often retain impressions 

 of muscle-scars and pallial lines. When such material is 

 not available, artificial moulds may be prepared, provided 

 the infilling material is compact The most efficient method 

 is that of " burning." In some cases, especially when the 

 matrix is only partly calcareous, the shells can be 

 actually calcined and removed by brushing. In others, 



