Nucleolus is the most prominent nuclear subcompartment in eukaryotes, where ribosomal RNA transcription and processing and their assembly into ribosomal subunits before exporting to the cytoplasm occur. The nucleolar evolution is one of the most important issues in consideration of the eukaryotic evolution. But until now, there are only few studies regarding to its evolution. We have first studied the genes involved in ribosome biogenesis in the anucleoate Giardia. Then, by comparative genomic analysis, we tried to provide some insights into the origin and the evolution of the nucleolus. Our results are as follows: 1) From anucleoate Giardia, we have cloned the gene of KRRlp homolog, which is presented in eukaryotes and involved in pre-rRNA processing. It is suggested that although Giardia doesn't possess nucleolus, its mechanism and pathway of ribosome biogenesis is similar to that of typical eukaryotes. In the typical eukaryotic cells with nucleoli, as one of the nucleolus proteins, KRRlp encoded by krrl gene is one of the proteins involved in pre-rRNA processing. Giardia'was once considered to be one of the most primitive eukaryotes. Lack of nucleolus is one of the supporting features. We have cloned the gene of KRR1P homolog in Giardia and demonstrated that it was actively transcribed. KRRlp physically and functionally interacts with a protein Krilp to form a complex, which is required for 40S ribosome biogenesis in the nucleolus in yeast. Our database searches indicated that Krilp homolog was also presented in G. lamhlia. It is suggested that the mechanism and pathway of ribosome biogenesis in Giardia is similar to that of typical eukaryotes. 2) Investigating genes involved in ribosome biogenesis in the anucleoate Giardia comprehensively, we found that there are 89 orthologs of the 129 ribosomal biogenesis proteins in Giardia, including those involved in rRNA methylation or pseudoiiridine and present in 90S, pre-40S and pre-60S particles. These data suggest the ribosome biogenesis system of Giardia is similar to that of typical eukaryotes, but probably simpler, and is different from that of prokaryotes. According to the result 1 and 2, we argued that although Giardia does not possess nucleolus, its mechanism and pathway of ribosome biogenesis is similar to that of typical eukaryotes. This may imply that the origin of ribosome biogenesis function is earlier than the formation of nucleolus in eukaryotes evolution. So it is possible that the real reason for the forming of nucleolar structure is for the newly uncovered, nontodMonal roles. But there is still another possibility that the anucleolate state is due to the reductive evolution during the parasite lifestyle of Giardia. If this is true, such reductive evolution is also an important phenomenon of adaptive evolution and deserves further study. 3) The proteins of the nucleolar proteome of the human, arabidopsis and yeast were characterized successively. Using these sequences to search KOG database, we found that about 74% human nucleolar proteins, 75% yeast nucleolus proteins and 84% arabidopsis nucleolus proteins are conserved in animal, fungi and plant. It indicated that most nucleolar proteins may originate earlier than higher eukaryotes divergence. There are 154 human proteins and 134 yeast proteins localized in both human, and yeast nucleolus. What makes -the difference of the protein number between the human and yeast is 10 eases of gene duplication in human. Similarly, we found there are 21 cases of gene duplication in humaa among the proteins, which are localized in both human and arabidopsis nucleolus. So we concluded that along with the evolution of eukaryotes., gene duplication is an important way of the nucleolus evolution, 442 human nucleolus proteins, which are also conserved in fungi and plant, were used to search the bacteria and archaea genomes. The results showed that there exist homologs in both genomes, but proteins only having homologs in archaea are more than those only having homologs in bacteria. When classifying these proteins according to their fimction annotations, we found that proteins, which only have homoles in archaea, were those involved in RNA modification transcription, and ribosomal proteins. Proteins only having homologs in bacteria were those such as RNA helicase and WD repeat proteins. So we argued that though the nucleolus has a complex origin, the main part of it might originate from archaea.