Many　previous　studies　have　shown　that　the　underestimation　of　the　plant　area　index　(PAI)　by　traditional　optical　instruments　is　mainly　due　to　the　clumping　effects　of　forest　canopies.　However,　the　evaluation　of　the　clumping　effects　of　canopy　(leaves,　stems,　branches,　fruits,　and　flowers)　and　woody　components　(stems　and　branches)　within　forest　canopies　is　still　a　challenging　task　for　current　indirect　instruments　and　algorithms.　In　this　study,　two　key　issues　of　the　gap　size　distribution　algorithm　(CC)　were　discussed　first,　such　as　the　stop　conditions　of　the　gap-removal　procedure　and　gap-removing　methods.　Four　types　of　gap-removing　schemes　were　compared　and　their　impacts　to　the　estimation　of　canopy　and　woody　components　clumping　indices　(${\Omega　-e}$　and　${\Omega　-w}$)　were　analyzed.　Obvious　discrepancies　were　found　between　the　estimates　of　${\Omega　-e}$　and　${\Omega　-w}$　from　the　four　gap-removing　schemes　at　each　zenith　angle.　A　new　gap-removal　scheme　was　proposed　subsequently.　After　that,　a　solution　was　presented　to　estimate　${\Omega　-e}$　and　${\Omega　-w}$　at　multiple　zenithal　and　azimuthal　directions　in　three　mature　Picea　crassifolia　forest　stands.　These　estimates　were　retrieved　based　on　multispectral　canopy　imager　(MCI)　equipment　and　CC　algorithm.　The　estimates　from　the　MCI　equipment　were　later　compared　with　those　from　the　TRAC　instrument.　Small　differences　were　found　between　the　${\Omega　-e}$　estimates　from　the　MCI　and　TRAC　instruments,　indicating　that　the　MCI　instrument　was　effective　for　evaluating　the　clumping　effects　of　forest　canopies　at　multiple　zenithal　and　azimuthal　directions.　Results　showed　that　both　of　${\Omega　-e}$　and　${\Omega　-w}$　change　with　the　zenithal　and　azimuthal　directions.　©　2008-2012　IEEE.