(,306 ) 



2. Mus. Vindob. ? ad., Rio Boraxudo, Dear Paraiiagm'i, Parana, December 

 1820. Natterer coll. Wiug 190 ; tail 170 mm. 



3. Mus. Viudob. c? ad., Rio Boraxndo, December 1820. Natterer coll. Wing 

 195 ; tail 174 mm. 



4. Mus. Viudob. ? , Rio Boraxudo, December 1820. Natterer coll. Wing 183 ; 

 tail IGO mm. 



(Nos. 2 to 4 types of C. heckelii Pelz.) 



5. Mus. Tring, No. 697, ? ad., Roca Nova, Serra do Mar, Parana, October 27, 

 19111. A. Robert coll. Wing 192 ; tail 166 mm. 



6. Mils. Tring, No. 718, ? ad., Roia Nova, Serra do Mar, Parana, November 4, 

 191)1. A. Robert coll. Wing 193; tail 165 mm. 



Besides the above specimens I have examined : 2 c? J, 1 ? , Ignape, S. Panlo ; 

 one adult, y])auema, 8. Panlo; 2 c?c?, 2 ??, Scaramuza ; 3 ??, Ytarare, 

 S. Paulo; 1 (?, Jaguaraiba, all collected by Natterer, except those from Ignajie, 

 which were obtained by Mr. Krone, and the one from Ypanema. I compared also 

 a series from Taqnara and Arroio Grande, Rio Grande do Snl, and two adults from 

 Blumeuan, Sauta Catharina (Mus. H. v. Berlepsch). 



The examination of this material proves, beyond doubt, that the shade of the 

 blue colour, on which the "species" C. heckelii aud C. inexpectatiis have been based, 

 is of no geographical significance, since we find, at the same localities, verditer-blue 

 {inexpectatiis} aud purjale blue {hechelii') specimens along with typical caeruleus. 

 This is especially illustrated by the series from Rio Grande do Snl • in Count 

 Berlepsch's collection. Three skins from Iguapd, and two others from Blumenau, 

 agree perfectly with the types of C heckelii, and I at first thought the latter 

 might be a difterent form confined to the coast region. Two ? ad. from the Serra 

 da Mar (almost topotypes of C. heckelii), however, do not belong to tliis purple-blue 

 form. One (No. 5) is, like the type of C. iiie.cpectatus, verditer-blue, the other 

 (No. 6) is typical caeruleus ! One adult from Ypanema, on the other hand, is 

 exactly intermediate in colour between the i)urplc-bluo {heckelii) and the caeruleus 

 phase, aud a female from Scaramuza has the lower [larts almost as purple as the 

 types of C heckelii 1 Moreover, the latter " form " has been lately recorded from 

 Paraguay ,-f the tj^pical locality of C. caeruleus I There is no constant diti'erence 

 in the colour of the throat or in the shape of the frontal feathers ; nor are the 

 dimensions diflerent. The types of C. heckelii being in very worn ])lumage, the 

 throat appears dull brownish, black, without any gloss. I may also mention that 

 the typical specimen of C. inexpectatus is an immature bird, with the under jiarts 

 mainly brownish grey, but some verditer-bhiish feathers are just coming out. 



I append the measurements of Natterer's series and some other specimens, to 

 show that there is no difierence in size. 



C. heckelii-^ha,se (purple-bine). 



1 ? ad., Tguap^. Wing 180 ; tail 152 mm. 



2 cJ ad. (sex ?), Iguape. Wing 184, 191 ; tail 150, 162 mm. 



Intermediate between heckelii and caeruleus. 

 Ad., Ypanema. Wing 192 ; tail 175 mm. 

 ? ad., Scaramuza, August 2, 1820. Natterer coll. Wing 192; tail 180 mm. 



' Cf. also Berlepsch, Zfitschr. gea. Ornitk. ii. p. 127. 

 f Oberholser, Proc. V. S. Nat. Mus. xxv. p. 140. 



