54 



spore group, each section, with or without a capillitium, has two orders, so that 

 we have here seven orders distributed over four groups ; in four of the orders a 

 capillitium is always present, in three it is absent. 



If we take each section and group in which there is more than one order, 

 we see that the principal features which are relied upon as distingruishing the 

 orders from each other are : — 



In the violet spored group with capillitium ; deposits of lime on the surface 

 or outside the sporangium in one order ; without deposits of lime on the surface 

 in the other orders. 



This is a technical distinction wliich will seldom come into requisition, 

 since the characters of the families in each order will soon be acquired, and after 

 all it will be discovered that practically the characters of the genera and families 

 are the most useful. 



In the coloured spore group with capillitium, one order has a columella, and 

 the other none. In the group without capillitium one order has a perforated or 

 reticulated inner wall to the peridium, which usually becomes exposed, and in 

 the other order there is no such inner wall. 



It would be tedious and unnecessary to proceed in this manner to analyse 

 the families and genera in each group, which can be done by anyone for himself 

 with the English translation before him. It will be found that in all genera 

 where a columella or capillitium is present, the distinctive features of the genus 

 are derived from the one or both ; and in genera, which possesses no capillitium, 

 that the distinctive features will be found in the character of one or both walls of 

 the peridium, or in the peridium in its entirety and its habit of development. 



If we take one family as an example, it will illustrate the mode in which 

 the principles are applied. In this family the threads of the capillitium are 

 thickened at intervals in the form of warts, spines, rings, half-rings, or reticu- 

 lations, and this is the Arcyriaceoe. 



In Lycagola several sporangia are combined into a globose complex or com- 

 pound sporangium. In the other genera the sporangia are simple. In Oligonema 

 the capillitium consists of single tubes. In Lachnoholus the network of the capil- 

 litium is grown to the whole walls of the sporangium. In Cornuvia the network 

 of the capillitium is quite free. In Arcyria the ends of the network are grown to 

 the base of the peridium. 



Thus it will be seen that certain characteristics of the capillitium constitute 

 the distinguishing features of the genera, and a similar method is adopted in 

 other families. 



A word or two as to the limits of species. This will probably be the first 

 point of divergence between Rostafinski and continental mycologists. There is 

 every probability that the whole classification will be universally accepted down 

 to genera, but when we come to the limits of species, it is more than probable that 

 a vast accession will soon be proposed. It must be conceded that Kostafiuski 



