In this thesis different magnetic nanoparticle systems have been investigated with the objective of finding materials most suited to self-controlled hyperthermia applications. This requires materials with Curie temperature close to the therapeutic limit of 42 - 47 °C together with large heat dissipation in RF magnetic fields. These two objectives are usually difficult to achieve in a given system. Therefore three different approaches have been used to address the problem; these include tuning exchange interactions, modifying magnetic anisotropy and reducing dipolar interactions between magnetic nanoparticles. Different nanoparticle systems viz. La1-xSrxMnO3 and mixed phase composites based on SrFe12O19 and MgFe2O4 and ZrO2 have been investigated in this context. In case of the strontium doped lanthanum manganite La1-xSrxMnO3 nanoparticle system, the exchange interactions and thereby the magnetic properties have been tuned by varying the Sr content x in the range 0.15 - 0.45. It was found that both magnetic and thermomagnetic behaviors are governed by the strontium content x. The saturation magnetization, coercivity and SAR vary non-monotonically with x. The measured SAR was found to be in close agreement with theoretically determined values obtained using the linear response theory (LRT). In the second approach using this system, the effect of particle size on magnetic anisotropy of La1-xSrxMnO3 with x lying in the range of 0.20 ≤ x ≤ 0.45 has been investigated. Magnetic properties such as saturation magnetization and Curie temperature were found to increase with the increase in particle size for each concentration. The measured SAR is maximum for particles lying in the range 25 – 30 nm for all values of x. Good agreement was found between the experimental and theoretically determined values of the SAR for samples lying in the single domain regime and having the largest anisotropy energies. It was therefore concluded that the effective anisotropy is the key parameter determining the SAR of in La1-xSrxMnO3 nanoparticles. Also, the LRT can be successfully used to calculate the SAR of these nanoparticles, provided they possess large enough effective anisotropies. Mixed phase composites based on magnetically hard SrFe12O19 and soft MgFe2O4 have been investigated by varying the weight percentage of the constituent phases whereby ZrO2 was used as a non-magnetic component. Room temperature magnetization measurements of the samples show significant variation in saturation magnetization, coercivity and remanence depending on the amount of the highly anisotropic Sr-hexaferrite phase. The composite samples show significant magnetothermia effect as opposed to pure SrFe12O19 in which no heating could be observed. This is due to the remarkable softening in the magnetic behaviour of pure SrFe12O19 upon addition of small amounts of the soft-magnetic MgFe2O4 and the non- magnetic ZrO2, making these composites suitable for magnetic hyperthermia. The effect of reducing dipolar interactions on the SAR was investigated in MgFe2O4 and ZrO2 composite nanoparticles with different weight percentages of ZrO2. The objective of introducing ZrO2, a biocompatible ceramic, was to prevent MgFe2O4 nanoparticles from aggregation and to reduce interparticle magnetic dipolar interactions in order to enhance the specific absorption rate (SAR). The blocking temperature and coercivity were significantly reduced in the composite samples by increasing the content of ZrO2 phase, indicating a decrease in interparticle interactions. This is an important finding from the point of view of biomedical applications, because ZrO2 in known to have low toxicity and high biocompatibility in comparison to that of ferrites. The reduced dipolar interactions were found to play a pivotal role in enhancing hyperthermia and we therefore, suggest the suitability of these composites as efficient mediators for magnetic hyperthermia.