"One group of animals was exposed to Digital Advanced Mobile Phone System (DAMPS)-type RFR, the other group to Global System for Mobile (GSM)-type RFR at a nominal average specific absorption rate of 0.5 W kg-1, 1.5 h day-1,for 5 days a week."

"Exposure to UV radiation The animals were exposed to UV radiation using lamps simulating the solar spectrum (Philips HP3136). The emission spectrum (see Kumlin et al, 1998b) and intensity of the lamps were measured with a erythema UV intensity and UV-A intensity meter (Solar Light Model 3D Solar Light Co., Inc., Philadelphia, PA, USA), and spectroradiometer (Macam Flexiscan,

Macam Photometrics Ltd., Livingston, UK). The dose of the UV radiation was 1.2 human minimum erythemal dose (MED), i.e. a dose of 240 J m{2, calculated according to the Commission Internationale de l’Eclairage (CIE) wavelength erythemal weighting function (McKinlay and Diffey 1987) (spatial variation of dose in a exposure cage within ¡10%, and the uncertainty of the measured values was estimated to be ¡10%). The UV exposure took place three times a week. The distance of the lamp from the animals was adjusted to reach the intended dose in 35 min. The back of each animal, including the controls, was shaved at least once a week with an electric clipper (Aesculap1 Favorita II, Tuttlingen, Germany). During the UV irradiations, the mice were kept in Macrolon III cages with a shallow wire lid preventing them from piling upon each other. To reduce the thermal load, the exposure cages were ventilated by a small electric fan."

"Exposure to RFR Three identical rectangular waveguide chambers made of aluminium were used for the RFR exposures: one for the sham-RFR exposure, one for the RFR (DAMPS) exposure and one for the RFR (GSM) exposure. For better control of absorption, the mice were kept in small cylindrical acrylic restrainers (32 mm i.d., length adjustable). The restrainers allowed some movement, but prevented the mice from aligning their longitudinal axis parallel to the electric field that would have increased the absorption significantly. The restrainers were kept by a Styrofoam holder at the centre of the cross-section of the waveguide with the animals’ longitudinal axis perpendicular to the electric field and to the direction of propagation of the RFR. Simultaneous exposure of 25 mice was possible in a chamber; each group of 50 animals was exposed in two consecutive daily sessions. The exposure order of the sessions was changed weekly. The exposure system was described in more detail by Heikkinen et al. (2001). For the RFR (DAMPS) group, a computercontrolled mobile phone for the DAMPS network was used as a signal source. The pulse duration was 6.67 ms, and the pulse frequency was 50 Hz. The phone operated at 849 MHz. Correspondingly, a modified computer-controlled phone for the GSM network, operating at 902.4 MHz, was used for the RFR (GSM) group. The pulse duration was 0.577 ms and pulse frequency 217 Hz. Speech was simulated by modulating the carriers with random sequence of bytes. The GSM signal was amplified by boosters manufactured for vehicle use, and the DAMPS signal by Mini-Circuits LZY-2-amplifier. Rohde & Schwarz NAS-Z7 power meters were coupled to the input and output of both exposure chambers for measurement of the input, reflected and output power. The output power was absorbed by a coaxial 10 W termination in both RFR-exposure chambers. The RFR-shamexposed group was kept in an unenergized waveguide chamber. The average specific absorption rate (SARave) was calculated from the measured powers and the mass of the mice: SARave~½ ð Þ 1{g=100 Pin{Pr{Pout =m, ð1Þ where Pin is the input power, Pr is the reflected power, Pout is the output power, g is the measured absorption of the Styrofoam holder as per cent of Pin, and m is the total mass of the 25 mice in each session. The nominal (time-averaged) whole-body average SARave was 0.5 W kg{1 in both the RFR (DAMPS) and RFR (GSM) groups resulting in whole-body average SARs of 1.5 and 4.0 W kg{1 during the pulses, respectively. The SAR was controlled daily by adjusting the input power. The mice were exposed to RFR for 1.5 h day{1, 5 days a week for 52 weeks. The dosimetric assessment performed at 902 MHz was described by Heikkinen et al. (2001). The maximum deviation of the SARave of the individual mice was estimated similarly at 849 MHz using 37-g phantoms instead of mice. The electric field measurements revealed no higher-order modes in the exposure chamber at 849 MHz. The previously reported maximum deviation of the SARave during exposure at 902 MHz (¡30% for a adult mouse) was found to be valid also at 849 MHz. Similarly to the previous study, the order of animals in the Styrofoam holder was randomized for each exposure session to ensure identical long-term average exposure. The maximum time-averaged local SAR in the skin (3.1 W kg{1 for a small mouse, 2.2 W kg{1 for an adult mouse) determined with FDTD calculations (Pitka¨aho et al. 1999) at 902 MHz may also be applied at 849 MHz. In the skin area where the UV-induced skin changes typically developed (an area of about 261 cm along the midline of the dorsum), the estimated average SAR values were about 0.3–0.4 W kg{1."

K2 mice (transgenic - overexpressing ODC), female

HAWC