1 \ f 7 Jt* 
'|jfl 
i i 
in about 10 days: Dept of Biology, University of Pennsylvania Philadelphia , Pa., 19104 
' 
HARVARD UNIVERSITY 
The Biological Laboratories 
16 DIVINITY AVENUE 
CAMBRIDGE 38, MASSACHUSETTS 
iijl 4 * •' -T / 14 July, 1964 
\ 
l V 
v. 
1 
41 
jYfj 1 1 I 
m. i } t . I ft 
ml 
By 
> t 
Dear Martin: " . i ’« • \ ■ J ‘ • 1 . • • ftW J 
Herewith the tape and Sonagraph tracings. I've copied the tape 
and fcetained a copy, so if this parcel is lost I'll eventaally get a query 
from you and be aole to start over# Also, I should be able to cheteb things 
1 or you without you risking mailing the master tape again# The copy, however, 
will be less perfect than the original, particular A ^in the frequencies over 
abtput 8 kc, since there is an ieevitable loss of fidelity in producing any 
copy. I’ m very sorry about the last sequence on your tape; I fear it(s largely 
destroyed# This happened as follows. Griffin's tape recorder has a speed error 
so I use rry Nagra to feed into the Sonagraph. The Nagra is a field machine 
and has a system of rollers designed to hold the tape steady in any position 
of the recorder - these don't function, however, as the free tail end of the 
tape passes through- they usually act to crimp or even break this tail, hence 
I put tails of leader tape on my material, and forgot I hadn't instructed you 
to do this. Fortunately, the last sequence was one you subsequently had me 
mark ’'skip" as it repeated earlier material, so the loss isn't as bad as it 
might have been. 
I 've made some interpretations right on the Sonagraph Iraces them- 
selves, particularly on the iirst ones. I think you're now experienced enough 
with these that I've given you sufficient information, but write if you don't 
understand sonetning. A lot oi echo shows up, usually fairly weakly, and it 
would help in Juture il you can deaden the walls and ceiling of your recording 
area. (Some of what I've thought to be echo may be a second animal in synchrony 
you'll be able to tell this by listening carefully to the tapes, I didn't ilways 
check this possibility; there are some beautiful cases of synchrony,E.g.#V) 
^ ^ ave used a number of different frequency and time scales. For every 
individual sonagraph make sure you note what I've marked. Many sections have 
been done over on different scales to clear up some points. As I explained in 
my earlier letter , to examine low frequencies I have had to condense time, so 
that you can't xalways determine the ihape of a note, but in most cases I've 
managed to get you some indication of thisone way or another. Things done at 
scales may appear to give different information for the reason that 
the sound filter being used has different charcfe&ristics (which in some cases 
marked down); in general the frequency width of a sound will appear to be 
greater when I've used a scale that analyses to high frequencies. Occasionally 
com P ensa ted ~or this by using an entirely" different filter (described as a 
narrow band pass" filter) which has other undesirable characteristics but which 
show you how harrow some of those broad black smudges should, in fact, be. 
.9 ul./ oi these will help you make your interpretations - if I were you I'd & 
go through the entire set before I sat to try to work on any one in 
detail. 
Let me know what you think of them, ^est regards to all. 
id 
