30 
As to its niclification, I may mention that in the Ruahine ranges I met with a breeding-pair of 
these birds late in December. The sudden disappearance of the female and the cautious demeanour 
of the male satisfied me that I was in the immediate vicinity of the nest ; but I nevertheless failed in 
my endeavours to find it. The bird resented my intrusion on its sanctum by a peculiar purr, not 
unlike the alarm-note of the American Cardinal ( Cardinalis virginianus), accompanied by a sudden 
spreading of the tail. 
A native once described the nest to me as being of large size and composed of moss, twigs, and 
dry leaves. He assured me that he had twice met with it in the high scrub near the Manawatu 
river, and that in both cases the nest contained two eggs. This was many years ago ; but that the 
account was reliable may be inferred from the fact, since ascertained, that this description applies 
very well to that of a closely allied species in the South Island. 
Although Turnagra has hitherto been placed among the Turdidse, the form is admittedly an 
aberrant one. Dr. Einsch has suggested the propriety of uniting it to Glaucopis, but I do not think 
this view has met with any acceptance or support. Fortunately I was able to bring with me to 
England a specimen in alcohol, which I forwarded to Dr. Gadow, of Cambridge, for anatomical study. 
After making an autopsy, with his accustomed care, he writes to me as follows : — “ I am sorry to say 
that the outcome of my investigation regarding Turnagra is not very striking. After all, you are 
quite right in your suggestions as to its position and affinities. The fact is, we know so little of the 
anatomy of the many birds belonging to the Timeliidse that comparison with these forms is almost 
out of the question. At any rate, it is satisfactory to know that there are not present any known 
characters to indicate other affinities, or to negative your suggestions.” 
Mr. Sharpe has placed Turnagra among his Timeliidse ; but I have decided to make it the type 
of a new family, Turnagridse, because the form seems to differ quite as much from typical Timelia as 
it does from Turdus. 
As it is important to place on permanent record the results of Dr. Gadow’s patient study of 
the subject, I shall here append his report in full, together with his detailed remarks on Glaucopis 
(referred to on page 4), in order to show that there is no relationship between these two forms, not- 
withstanding the similarity of some of the external characters : — 
“ Tckn'agra. — S tomach quadrangular, flattened, very muscular. Crop absent. Tongue fleshy, with a few short bristles on 
the sides near the tip. Intestinal convolutions Thrush-like, certainly not Corvine, with decided graminivorous adaptation. 
Syrinx muscles acromyodean. Pterylosis agrees with Nitzsch’s Subulirostres s. Canora;. Ten primaries ; terminal (or first) long ; 
tip of wing formed by third to seventh ; sixth longest. Nine secondaries. Twelve tail-feathers. Metatarsus like that of 
Thrushes or Sylvise. Sternum and shoulder- girdle agree with many birds : Struthidea, Oraucalus, Strepera, Ptilonorhynchus, 
Turdus (i. e. all alike). Conclusion : After examination of the digestive apparatus, the pelvic nerve-plexus, the skeloton, and the 
pterylosis, I feel inclined to put Turnagra with the wide and ill-defined group of Timeliidse. Turnagra is certainly neither 
Corvine nor Fringilline, and it is in fact a member, of the Southern (Indian-Australian) mass of Thrush-like birds. Its bill 
and certain modifications of its digestive apparatus seem to show that this bird is a Thrush with graminivorous propensities. I 
would put it into Sharpe’s subfamily Ptilonorhynchinae, to which PEluroedus belongs, hut unfortunately Ptilonorhi/nchus itself is 
very different from Timeliida3 in its pterylosis.” 
GiATrcoms. — After examination of the skeleton I am satisfied that Glaucopis comes nearest to the Gorvida.. Tho skuXl , 
although in general configuration and beak very similar to that of Struthidea, differs from tho latter. Barring tho peculiar 
lacrymals,it agrees with Ptilonorhynchus, also with Strepera, and, more remotely, with Paradisea. No agreement with Grctucalus. 
Comparison with Heteralocha and Sturnus is not possible. Skull, consequently, agrees with Ptilonorhynchus and Strepera. 
Sternum : agrees most with that of Strepera, far less with Graucalus, Struthidea, Paradisea. Ptilonorhynchus disagrees in 
clavicles, like Heteralocha and Sturnus. Pelvis and sacrum : agrees with Graucalus, Heteralocha, and Ptilonorhynchus, also with 
Strepera, Paradisea, and Struthidea. Metatarsal scutes : agree most with Heteralocha ; through the fusing condition in which the 
scutes are, very much with Ptilonorhynchus and Gymnorhina. Hyoul hones : Corvid®. Pterylosis : Strepera and Ptilonorhynchus, 
hut the latter has considerably more remiges. Conclusion: Glaucopis is nearly allied to the Austroeoraces. It agrees best with 
Strepera (Gynmorhinince in general), and shows some considerable similarity in structure with Ptilonorhynchus. Struthidea 
agrees with Glaucopis by far less than you might perhaps suppose, and Graucalus is still further removed. Heteralocha is an 
nnmistakablo Starling form, and has little of importance in common with Glaucopis .” 
