4 



576 Goodlatte: The anatomy of Parosela spinosa 



" Der Sitz der Korkbildung ist librigens fur jeden Species, ja man 

 kann sagen Gattung, constant." Solereder, on the other hand, 

 quoting as authorities Sanio, J. E. Weiss, and Douliot, on cork 

 formation in the Papihonaceae, says that the position of the cork- 

 forming layer differs not only within the limits of a genus but also 

 '^ bemerkenswert ist, dass zuvveilen bei derselben Art der Ort der 

 Korkentstehung zwischen der zweiten und sechsten Zellschicht 

 der primaren Rinde wechsein kann." * Weyland finds that in 

 Dalea^ \\\ the genus as a whole the cork is formed in the second 

 to sixth cell-row of the cortex. In the plant under discussion it 

 is formed, however, on the very inner row of the primary cortex, 



ft 



immediately outside the primary stereome bundles. This would 

 be about the ninth or tenth row, as the number of rows of palisade 

 cells is variable. Hence there is considerable variability in the' 

 genus in this respect (figure 4), 



A cross-section of a young stem shows, first, the epidermis, 

 with a layer beneath it of cells containing rhomboidal crystals ; 

 then five or six rows of palisade cells, many of them containing 



■ 



rod-shaped crystals ; then, a layer of parenchyma cells, with rhom- 

 boidal crystals ; next, the row of collateral bundles and a cambial 

 ring, rather ill-defined, each bundle being accompanied on the side 

 toward the periphery of the stem by a strand of primary stereome 

 tissue. In the center, of course, are the pith cells, with rather 

 thickened walls, some of them transformed into tracheidal paren- 

 chyma with simple pores, and many of them, especially those 

 nearest the wood, containing rhomboidal crystals, sometimes in 

 the cell-cavity, sometimes imbedded in the wall. The only inter- 

 ruption in the simple regular character of the stem structure is the 

 occasional appearance of a few cauline bundles, of the collateral 



type, accompanied by a stereome strand outside the leptome 

 (figure 5). 



the young stem develops, the layer of cutin, as has been 

 remarked, becomes more strongly developed. When the cortex 

 is supplanted by cork cells the outer ones of these have cutlnlzed 

 walls, the inner (phellogen) layer retaining unmodified cellulose 

 walls. Traces of suberin appear in a few cells of the hadrome, 



As 



*Anat. Dicot. 313. 



