134 PROCEEDINGS OF THE NATIONAL MUSEUM. vol. xxxv. 



are well known and the specimen figured on Plate XVIII may be 

 taken as an illustration of this particular phase of the subject. 



In the first instance mentioned, the original cavity in the rock 

 may have been left by the removal of a fossil by solution, but just as 

 frequently the fossil itself furnishes the cavity. Figs. 3 to 8 on Plate 

 XIX are from photographs of some common Upper Ordovician fos- 

 sils well illustrating this occurrence. The large brachiopod, Platy- 

 strophia lynx (figs. 3 and 5), was evidently buried without being filled 

 with mud or other materials. Its hollow interior then afforded an 

 excellent place for the deposition of crystalline material. Only a 

 portion of the interior of the smaller brachiopod, Rhynchotrema 

 capax, figured on the same plate (figs. 6-8), was left unfilled with 

 clay, and a correspondingly small geode resulted. What may be 

 called a compound or, better, multiplex geode, results from the de- 

 position of crystals in the hollow chambers of cephalopods. The 

 Orthoceras, figured on Plate XIX, is an example of this on a small 

 scale, but larger specimens are not infrequent. In this group of 

 animals the shell, when intact, is less likely to become filled with 

 clay, since the only means of ingress to the interior of the camerated 

 portion is through the breakage of the walls of the camerae or of the 

 siphuncle. No resistance is offered to the passage of water through 

 the empty shell, and, as a result, almost every cephalopod in certain 

 strata can be rated also as a geode. In all of this second group of 

 geodes the crystalline material is calcite, usually in the form of dog- 

 tooth spar. The most striking and best-known geodes, however, are 

 siliceous, and it is to this class that the writer wishes to call special 

 attention. The examples referred to previously, as well as those that 

 follow, are of such common occurrence that the facts here presented 

 must have been noted by many observers. It was therefore with some 

 hesitancy that the present paper was undertaken. A search of the 

 literature, however, revealed little published information upon the 

 subject. Doubtless references to geodes and geodization exist, but 

 they are, apparently, as in the case of Professor Shaler's work re- 

 ferred to below, hidden under other titles. 



Among the many problems of silicification the one of most interest 

 to the paleontologist is unquestionably that bearing upon the re- 

 placement of fossils. Shells and other fossil forms entirely replaced 

 by silica are of common occurrence in the cherty or siliceous debris re- 

 sulting from the waste of limestones. These siliceous pseudomorphs, 

 as they are called, often preserve the original shape and markings 

 of the fossil form so perfectly that the peculiar interest of such a 

 process to the paleontologist is obvious. Indeed, some very interest- 

 ing faunas of geological time would be practically unknown if it 

 had been necessary to work out the individual species from the hard, 

 refractory limestone containing them. As an example of this, the 



