The　transcranial　magnetic　stimulation　(TMS)　we　designed　was　first　measured　in　terms　of　its　magnetic　field　intensity　distribution　regarding　three　aspects,　i.e.　the　distribution　of　magnetic　field　intensity　in　axial　direction,　in　radial　direction　and　around　the　circle　which　held　the　maximal　magnetic　field　intensity.　Then,　it　was　applied　into　a　therapeutic　rehabilitation　experiment　of　heroine　addicts.　Before　TMS　treatment,　the　electroencephalogram　(EEG)　results　of　patients,　who　were　undergoing　visual　stimulation　with　different　drug　pictures,　were　recorded.　These　pictures　were　given　to　the　patients　after　treatment　once　again　in　the　same　order,　and　then　the　EEG　results　were　also　recorded　to　make　a　comparison　with　the　former　ones.　The　two　groups　of　data　were　processed　in　a　certain　procedure　first,　and　then　the　obtained　N270　amplitudes　and　latencies　were　compared　and　analyzed.　The　findings　showed　that　the　N270　amplitude　of　the　drug　addicts　decreased　after　magnetic　stimulation,　and　the　latency　was　shortened.　There　were　also　brain　functional　changes　in　the　abnormal　encephalic　regions.　And　it　was　in　the　frontal　lobe,　the　temporal　lobe,　the　temporal　lobe　and　the　occipital　lobe　region　where　most　differential　electrodes　appeared.　We　concluded　accordingly　that　all　of　these　above-mentioned　brain　regions　of　the　patients　were　possibly　injured　by　heroin　and　functions　of　these　parts　were　abnormal.　However,　through　a　course　of　TMS　treatment,　the　N270　amplitude　and　latency　got　closer　to　the　normal　value,　which　suggests　that　the　TMS　contributes　to　the　recovery　of　heroin　addiction.