Time is a subject that has been discussed and researched on since ancient. Scientist and other more and more have studious of understood the time while time and science progressed. Time is a human construct and useful for describing detail of objects. By use of time, we can record and memorize our facts and experiences that happened late. With due attention to these capabilities, predict some events that expect occur on the future. (Temporal Mining) Thus we reach these concepts only when have been an explicitly definition of time. On the other hand Conventional database systems do not offer the possibility of dealing with time varying data. The content of a database represents a snapshot of the reality in that only the current data are recorded, without the possibility of maintaining the complete history of data over time. Numerous proposals for extending the relational data model to incorporate the temporal dimension of data have appeared in the past several years. While most of these have been historical databases, incorporating in some fashion a valid time dimension to the data model and the query languages, others have been rollback databases, incorporating a transaction time dimension, or bitemporal databases, incorporating both of these temporal dimensions. A bitemporal database is a database with exactly one system supported valid time and exactly one system-supported transaction time. These two temporal dimensions to data are the valid time of a fact as the time when the fact is true in the modeled reality, and the transaction time of a database fact as the time when the fact is stored in the database. We analyze the interrelationships among these two dimensions and a third dimension which we call the reference time. The reference time provides a means to refer to the state of the database as it would appear to an observer looking at the database at that time. Most object-oriented systems owe their origins to the programming language Smalltalk. Objects, the basic data constructs used in these systems, have proven to be both a powerful and flexible modeling construct. The power of objects arises in part from their ability to encapsulate both structure and behavior. The flexibility with which objects can be used as modeling constructs is due to the sets of data types that can be combined to define objects, the ability to nest the structure of objects And The ability to encapsulate operations or methods on these objects in the manner of data types. In the object oriented approaches, temporal dimensions can be defined at the object level like as TMAD, OSAM/T or at the attribute level like as Keywords: Object Oriented Database, Bi-Temporal Database, Snapshot Database, Time Point, Time Interval, Valid Time, Transaction Time