Avian flight cannot fail to impress; from the huge migratory distances covered by albatrosses to the dexterity shown by a feeding hummingbird, the performance of birds in flight is remarkable. Until now research into free flight (i.e. not in a wind tunnel or other artificial environment) has been limited to observations from the ground. Here I use a collection of novel techniques, based on the use of onboard instrumentation carried by the bird, to explore if and how this performance might be underpinned by their flexible flight configuration. In the Introduction (chapter 1) to the thesis I investigate previous work into the stability and control of birds in flight. In chapter 2 I investigate a selection of manoeuvres seen commonly in flight, and describe the ways in which they exploit the flexible configuration of the eagle. Then (chapter 3), using an Inertial Measurement Unit (IMU) containing an integrated camera and Pilot-tube I measure the forces acting on the body of the bird in flight, the airspeed of the bird, and using custom-written software, I extract the configuration of the tail from the video. Using the measured configuration of the tail I estimate the lift generated by the tail according to a series of simple aerodynamic models. These are good predictors of the variation in the normal load factor acting on the bird. This suggests that the tail of the eagle is used primarily in soaring flight to balance the bird along the pitch axis. In chapter 4 I further investigate the configuration of the tail, I find that the spread and angle of attack of the tail covary, but that the twist of the tail is adjusted independently. In chapter 5 I explore one manoeuvre, the wing tuck, in more detail. With reference to a 'mean wing tuck' of the key variables I suggest that it is a response to a drop in wing loading, which suggests that it may be a response to atmospheric turbulence. I then investigate the frequency of wing tucking and our principal finding is that it is increased on days when the wind speed is greater, further suggesting that it is a response to atmospheric turbulence. Finally in the Discussion (chapter 6) I summarise the thesis. I also consider future avenues for research into the control and stability of avian flight and discuss some of the limitations of the methods used in this thesis.

The Golden Eagle Aquila chrysaetos canadensis Linn. is facing extinction in much of its range. In the Western Hemisphere its numbers are restricted primarily to the mountainous regions west of the 99th meridian, from extreme northern Alaska to northern Mexico (Hobbie and Cade, 1962) and from timberline in Colorado to below sea level in Death Valley, California (Summer, 1929a).

The large size, extensive range, and irnporta.nt corrnnunity role of the Golden Eagle (Aguila chrysaetos L.) and the Bald Eagle (Haliae-etus leucocephalus L.) attract considerable scientific interest. Sorne aspects of their life history have been investigated and reported in detail, but their rapidly declining numbers (Sprunt and Ligas 1966, Cottam 1961) necessitate new investigations into all phases of their life history and particularly into population dynamics.