16 PAL^ONTOLOGICAL BOTANY. 



sufficient to maintain a very large object under the action of the 

 slitting-plate. By this method no time is lost ; the adhesion 

 is made instantaneously, and as quickly broken by opening a 

 small screw, to admit air between the glass plate and the 

 chuck, when the object is immediately released. Care must 

 be taken, in stretching the india-rubber over the face of the 

 chuck, to make it very equal in its distribution, and as thin 

 as is consistent with strength. When this material is obtained 

 from the shops, it presents a series of slight grooves, and is 

 rather hard for our purpose. It ought, therefore, to be 

 slightly heated, which renders it soft and pliant, and in this 

 state should now be stretched over the chuck, and a piece of 

 soft copper wire tied round it, a slight groove being cut in the 

 periphery of the chuck to detain the wire in its place. 

 When by use the surface of the india-rubber becomes flat, 

 smooth, and free from the grooves which at first mar its use- 

 fulness, a specimen may be slit of many square inches, with- 

 out resort being had to another exhaustion by the syringe. 

 But when a large, hard, siliceous object has to be slit, it is 

 well for the sake of safety to try the syringe piston, and 

 observe if it returns forcibly to the bottom of the cylinder, 

 which cAddences the good condition of the vacuum of the 

 chuck. 



After the operation of slitting, the plate must be removed 

 from the spindle of the lathe, and the flat lead lap substituted. 

 The pneumatic chuck is now^ to be reversed, and the specimen 

 placed in contact with the grinder. By giving a slightly 

 tortuous motion to the specimen, that is, using the motion of 

 the various joints, the object is ground perfectly flat when 

 the length of both arms of the joints is perfectly equal. 

 Should the leg of the first joint on the right-hand side be the 

 longer, the specimen will be ground hollow ; if shorter, it 

 will be ground convex. But if, as before stated, they are of 

 equal length, a perfectly parallel surface will be obtained. 



