192 



The cells iiiakiii,i;- up these disks are lar.ue. irregular in outline, very 

 thiek-walled. li.ijiiitied. and oontaiu stareh in winter. The cells tilling- the 

 interspaces are small, regular, very thin-walled, unlignitied and empty. 



The formation of the disks takes i)lace at a very early stage in the 

 growth of the twig; they may l)e seen just back of the growing point in 

 Fig. 1. Avhich is a longitiidinal section through a yoiuig twig of Liri- 

 odniilnm. 



The genus Magiioliri presents some interesting moditications of this 

 type. The genus has been descrilied as always having these partitions in 

 the pith: liut. several have pointed <>ul that this statenu'nt is incorrect. 

 In the examination of the American and some of the Asiatic species, I 

 have foxuid only two. .1/. \ irniniuiio and .1/. firlUhi. in which the fully 

 developed disks (HH'uri'ed. In all the other species oxaniined. cells of the 

 sort described as making up the disks occiu'red scattered singly or in 

 small groups tbroughout the iiith. P.aillon. in his Natural History of 

 riants. says of this: "In the rapidly developed shoots of some Magno- 

 lias we have seen these septa reduced to a single cell, nearly central, on 

 which all the surrounding cells of the ordinary parenchyma abut liy one 

 end. bent, or drawn out in a quite iicculiar fashion." 



In I''ig. ■_', wliich is a longilndiujit sedidn (if a twig of I/, trijxtdhi. 

 tJiesr scattered grouiis of cells are slmwn: and. Fig. ."! sliows tlie same 

 kind of cells in a cross-sectii»n of a twig of tlie same species. 



In Asiiiiiiiii the disks seem to lie made uji of more i-egidar ;iiid thicker- 

 walled cells tlian are found in Mdiinulid and /Jriuilciidrdii. 



In the slender woody twigs of Xiissii. vei-y strongly (h'velo]ieil disks 

 were found, stronger in fact than in any other case examined. 

 Finicfiiiii (if jiitli (if this lijpv: — 



Xii satisfactory explanation of the function of tliis type of jiitli has 

 been offered. From superlicial examination, the suggestion th.at its func- 

 tion was on(> of mechanical sui)port would seem reasonable; but. the fact 

 tliai the most strongly develoiied diaphragms were found in the strong: 

 and slender twigs of Xi/ssa, while the thick MmiuoVrt twigs with their 

 relatively large pith showed the weakest development of tiiis tyjie. seems 

 to indicate that the suggestion of mechanical support is not ;i sutlh-ient 

 explanation of their function. 



The second type of ])ith has often been mentioned and figured in 

 species of -lufiUnix. I have also studied it in I'tcro-urnn. Ccltis. Mohro- 

 (IcjidroH {lldlcsim. I'drsythid riridissinid. JdsiiiiiiKin si)cci(s. J'diihnniid. and 



