PASSEROIDE^E. 481 



or birds of prey. The clayey mud bakes almost to brick, and it is no easy matter 

 even to break out a hole large enough to extract the eggs. The nests are rather less 

 than a foot in greatest diameter, and though the eggs are not visible from the entrance 

 the common statement that there is an ' ante-chamber ' to the nest seems to me not 

 quite accurate. The nest is built very much like a spiral shell, and if one could re- 

 move the inner whorls from such a shell as Ampullaria he would have quite a fair 

 miniature of the hornero's nest. The eggs are seldom more than three in number, and 

 are originally pure white, but being laid directly on the muddy floor of the nest they 

 soon acquire about the same color. I have taken them from September 16 until Jan- 

 uary 15, but the larger number are, I think, laid during October." 



Finally, we shall have to copy Mr. Gibson's observations on the curious burrowing 

 habits of the terrestrial and lark-like Geositta cunicularia, made in Buenos Aires : 

 " This is a most common bird in the camps or plains, to which it is confined, though 

 we may notice it about the borders of any wood also, should there be a biscachero 

 (colony of La-gostomus trichodactylus) situated near. On the 16th of August I have 

 seen the excavation of a nest begun. This (with us, at least) is invariably situated in 

 the brow of one of the burrows of a biscachero ; and as a new one is made every year, 

 it is often difficult to tell which hole to open up. The passage varies in length from 

 two to four feet, with a slight downward tendency, and terminates in a cavity shaped 

 like a cocoa-nut, but a little larger. This is prepared for the reception of the eggs by 

 a pile or cushion of soft dry grass, sometimes very sparse. The slope of the passage 

 I have always found to be correctly calculated, so as not to disconcert the bird by a 

 possible emergence into the open air, should the ground fall away behind the burrow. 

 The bird sits close, and may often be excavated with the nest. Three is the general 

 number of eggs laid. The color is white." 



As the Passeres form the largest order among existing birds, so are the PASSEROI- 

 DE^ the largest super-family, agreeing exactly with the Acromyodi normales of Gar- 

 rod, the Oscines of Keyserling and Blasius, Mdller, Cabanis, and others, and the 

 Laminiplanteres of Sundevall, plus the larks. So far the harmony amongst systematists 

 is great and unusual. But when we come to subdivide this enormous group, the una- 

 nimity suddenly comes to an end, and, to use Professor Parker's words, " A hundred 

 classifiers, a hundred so-called systems ! " 



Having the task now before us of selecting one of these hundred systems, or, if 

 they do not suit us, to propose number one hundred and one, at least for our own use, 

 we will have to follow the principle which has guided us so far, viz., to begin with 

 the most generalized forms, proceeding towards the most specialized ones, whenever the 

 present status of the science enables us to draw reliable conclusions in that respect. 

 Our view in regard to the present super-family is fairly expressed by the following 

 quotation from Professor Newton : " Hard though it be to find definitions for the 

 several groups of Oscines, whether we make them more or fewer, it is by no means so 

 hard, if we go the right way to work, to determine which of them is the highest [most 

 specialized], and, possibly, which of them is the lowest [most generalized]." But it 

 must at the same time be borne in mind that a linear arrangement is as impossible in 

 this as in any other polymorphous group ; an attempt to arrange all the forms belong- 

 ing here, in one natural series, is bound to fail, because it is against Nature. 



In most modern systems we find that the 'highest' position has been granted the 

 Turdidee, partly because some of that family are among the best songsters, partly 

 because they possess the most specialized tarsus, the scales, both in front and behind, 



VOL. IV. 31 



