PLATYSTERA. 189 



much closer than either of the three already pointed 

 out to Conopophaga, and yet preserving the general 

 characters of its own group. The importance of 

 such a form is very great, for it not only gives us a 

 repetition, even in a sub-genus, of the fissirostral 

 type ; but it brings Platystera in immediate con- 

 junction with Conopophaga, the first group in the 

 circle we have been so long in tracing, and the last 

 to which we again return. Whether this Platys- 

 tera, which we had named provisionally brevicau- 

 da, still exists in the National Museum, we know 

 not ; but at the time the above note was made upon 

 it, that law of variation which renders Platyrhyn- 

 chus the raptorial type of the genus Todus, had not 

 been discovered; and this explains a part of the 

 foregoing memorandum. 



We will now compare the species representing 

 the types of form in the sub-genera of Todus and 

 Platystera, and ascertain in what way they resem- 

 ble each other. 



Species of Pla- g e <e8otTodvt 



leSg'tTpes AN ^O<>ICAL CHARACTERS. representing 

 of form. types of form. 



P. lonyipes Legs and tail very long T. platycirctu. 



( Bill much depressed, the ) , . 

 P. pectoralis.... \ tip abruptly and strongly \- *' i 1 "*?**- 



C hooked. ) lus ' 



P. perspicillata....Al\ the toes syndactyle T. viridis. 



P. brevicauda Tail remarkably short T. meyacephalut. 



We were perplexed, for many years, in endea- 

 vouring to make out why it so frequently happened 

 that in very small groups, like sub-genera, there are 



