THE gyr falcon (Falco rusticolus) is the largest falcon species in the world, and inhabits Arctic regions of North America, Greenland, Iceland and Eurasia. The species is considered polytypic, with four races, F r rusticolus in Europe, F r obsoletus in Asia and parts of North America, F r candicans from the high Arctic of North America and Greenland, and F r islandicus from Iceland. The gyr falcon is a medium-sized bird of prey, measuring 48 to 60 cm in length and with a wingspan of 120 to 135 cm. The bodyweight of males ranges from 961 g to 1321 g, and of females from 1262 g to 2100 g (del Hoyo and others 1994). The gyr falcon is one of the species used in the traditional sport of falconry in the Middle East. However, captive-bred or wild-caught gyr falcons have never been in great demand in the region, as these highly expensive birds appear to be more susceptible to disease, notably bumblefoot and aspergillosis (Samour 2000). It has also been suggested that gyr falcons seldom perform well and prove difficult to train and to engage large quarry (Upton 2002). Despite this, the number of captive-bred and wild-caught gyr falcons in captivity appears to be steadily increasing throughout the Middle East. For instance, at the authors’ hospital, only 15 gyr falcons were examined during the 1999/2000 season (September to March); this number increased to 37 during the 2000/01 season, 65 during the 2001/02 season and 71 during the 2002/03 season. Due to the increasing number of gyr falcons kept in captivity throughout the Middle East, it has become important to establish normal reference laboratory values in order to assess health and disease in individuals. Normal haematology reference values for captive-bred gyr falcons have recently been published by Wernery and others (2004). This short communication describes the results of a haematology study carried out in recently caught free-living gyr falcons used in falconry. Blood samples were collected from 25 adult (over nine months of age) female gyr falcons presented for general examination to the Falcon Specialist Hospital and Research Institute of the Fahad bin Sultan Falcon Center, Riyadh, during the 2003/04 season. All falcons had been wild-caught and were examined within two weeks of arriving in Saudi Arabia; it was estimated that the falcons had been in captivity for less than one month. All the birds were subjected to a systematic clinical assessment and declared clinically normal. The clinical assessment included a full physical examination, stress test and ventrodorsal and laterolateral radiographs. A blood sample was obtained from a basilic vein from each bird using a 2 ml disposable syringe and a 23 G, 5/8” disposable needle. The samples were collected while the falcons were under isoflurane anaesthesia (Isoflo; Mallinckrodt Veterinary) in oxygen, before the radiological examination. After collection, 0·3 to 0·5 ml of the blood was mixed with EDTA at 1·5 mg/ml blood in commercially available storage tubes. Haematological analyses were undertaken within 10 to 20 minutes after sample collection. Red blood cells (RBC) were counted manually by mixing 20 μl of blood into 4 ml of formol citrate solution and examining a sample by phase-contrast microscopy (Dacie and Lewis 1995) using an improved Neubauer double cell clearsight haemocytometer (Weber Scientific International). Haemoglobin was estimated as azidemethaemoglobin using a dedicated haemoglobinometer system (HemoCue). This system includes reagent-preloaded microcuvettes and a photometer. The readings are carried out at 570 nm and 880 nm to compensate for turbidity within the sample. For the estimation of haematocrit, plain microcapillary tubes (Hawksley & Sons) were filled to approximately three-quarters of the length with blood, and after one end was sealed using a commercially available sealing compound (Hawksley & Sons) the tubes were centrifuged at 10,000 g for five minutes using a microhaematocrit centrifuge (Hawksley & Sons). Haematocrit readings were carried out using a manual microhaematocrit reader (Hawksley & Sons). Fibrinogen was estimated as protein precipitated using the same microcapillary tubes used for the measurements of haematocrit (Millar and others 1971). After the estimation of haematocrit, the microcapillary tubes were heated at 56°C for three minutes in a waterbath and then recentrifuged at 10,000 g for five minutes. The microcapillary tubes were then placed on to a modified microscope slide and measurements were taken at the interface of the white cell-thrombocyte layer, at the top of the protein layer and plasma column, using a light microscope fitted with a 16 mm objective using the microscope stage Vernier and a micrometer eyepiece (Hawkey and Samour 1988). The concentration of fibrinogen was calculated from these measurements. The mean corpuscular volume (MCV), mean corpuscular haemoglobin (MCH) and mean corpuscular haemoglobin concentration (MCHC) of the RBCs were calculated using standard formulae (Dacie and Lewis 1995). White blood cells (WBCs) were counted manually by mixing 100 μl of blood into 1·9 ml of 1 per cent ammonium oxalate solution and examining by phase-contrast microscopy (Hawkey and Samour 1988) using an improved Neubauer double cell clearsight haemocytometer. Differential WBC and thrombocyte counts, and assessment of the morphology of white and red cells, were carried out in blood films fixed in acetone-free absolute methanol for five minutes and stained using a modified Wright-Giemsa staining procedure. The working stain was prepared using 3 g Wright stain powder, 0·3 g Giemsa stain powder, 5 ml glycerol and absolute methanol to 1000 ml. The staining protocol was as previously described by Campbell (1992). Blood films were examined using a x 100 oil immersion lens. White and red cells were measured under a x 100 oil immersion lens on the same blood films using a micrometer eyepiece. The results of the morphometric study of the RBCs, WBCs and thrombocytes in the gyr falcons are shown in Table 1. The RBCs were elongated, oval-shaped cells, mean (se) 14·82 (0·07) x 7·21 (0·04) μm, with slightly basophilic (blue-grey) cytoplasm containing a relatively large nucleus with evenly distributed chromatin clusters. Scanty immature red cells at different stages of development were observed in all films examined. One or two red cells without nuclei, or erythroplastids, were also observed in most films. The heterophils were large round cells, 12·72 (0·11) μm, with colourless cytoplasm containing a bilobed nucleus. The cytoplasm contained Veterinary Record (2005) 157, 844-847