82 Systematic Arrangement. 



CHAPTER XXI. 



SYSTEMATIC ARRANGEMENT AND TECHNICAL DESCRIPTIONS. 



The different Rusts found in Australia are here named, described, and 

 systematically arranged. The names are necessary to distinguish one from 

 another, for, as George Eliot happily puts it " The mere fact of naming 

 an object tends to give definiteness to our conception of it. We have then 

 a sign which at once calls up in our minds the distinctive qualities wihich 

 mark out for us that particular object from all others." fT-he descriptions 

 are necessarily technical, and give those characters which enable the species 

 to be discriminated from others, with the help of the illustrations. When 

 the described 1 stage of any rust is enclosed in square brackets, this indicates 

 that it has not been found in Australia. The systematic arrangement 

 deals with the nine genera at present known, and arranges them according 

 to their natural affinities. Taking a general view of the entire order, 

 the following scheme of classification will be adopted, mainlv based upon 

 that of Engler and Prantl, in their Die natiirlichen Pflanzenfamilien: 



Order UREDINEAE, Tul. 



Fungi parasitic on higher plants and developing in the interior of their 

 hosts, a filiform, branching, septate mycelium. Spores arising terminally 

 or laterally from erect, transversely divided, crowded hyphae, and usually 

 of more than one kind. Teleutospores germinating by a short promy- 

 celium. The order may be grouped in the following four families, of which 

 the third is not represented in Australia: 



Teleutospores stalked, single, in groups or rows, or several cells in a 

 head. Fam i. Pucciniaceae. 



Teleutospores sessile, in columnar or filiform masses. Fam. 2. Cro- 

 nartiaceae. 



Teleutospores sessile or stalked, in one or two-layered waxy masses. 

 Fam. 3. Coleosporiaceae. 



Teleutospores sessile, in flattened one-layered masses, or loose in the 

 tissues of their host. Fam. 4. Melampsoraceae. 



On account of their economic importance, the Graminaceous rusts are 

 always treated first and the others are likewise grouped together under 

 their respective families of host-plants, the order generally followed being 

 that of Baron von Mueller's Systematic Census of Australian Plants, start- 

 ing from the Grasses, and ascending to the higher forms. 



FAM. i. PUCCINIACEAE. 



Since the great majority of our Australian Rusts belong to this family, 



be treated at greater length than any of the others, and it will be 



nterestmg to trace (the different forms which the teleutospore assumes in 



the different genera. There are a number of genera based upon very slight 



ifTerentiating characters, and it is not always easy to settle whether they 



etamed or rejected, but I have given all those which are clearly 



tmct. It is a moot point whether such a genus as Diorchidium Kalch.. 



lould be retained, where the teleutospore is two-celled, as in Puccima 



only the septum is longitudinal instead of transverse. There are several 



j of Puccima in which there is a tendency to an oblique, and even a 



longitudinal septum, as well as transverse, but this genus is retained for 



those species in which the spore is distinctly divided longitudinally with a 



single germ-pore at the apex of each. 



