When pollen grains are extracted from their sediment matrix, the different types of plants which grew in the area at the time the sediment was laid down can be recognized from the distinctive characteristics of their pollen grains. The _pollen analyst_ has the job of interpreting this information to discover the nature of the vegetation patterns in the past and the climates associated with them. Through dating those climates, the artifacts with which the pollen sample was associated are dated. Some archaeological sites do not allow dating by any of these methods. The farmer plowing a field may pick up an arrowhead and wonder how old it is. Can the archaeologist date the arrowhead? The answer is a qualified yes. What the archaeologist will try to do is give an educated guess as to the age of the arrowhead. He might, for example, know that arrowheads of that type are found at a site which is dated about A.D. 1000 by various methods. By correlation, he could apply that date. If the arrowhead is not of a type which has been reliably dated, the archaeologist may rely upon the method of dating known as _seriation_. This method depends on the second assumption basic to the principle of stratigraphy, that similar objects tend to be about the same age. The farmer’s arrowhead may not be exactly like any which has ever been discovered, but it probably will be more like some known ones than others. Observing the general style of the arrowhead and the way in which it is made, the archaeologist can make a pretty good estimate of when it was made. Let us assume that the archaeologist goes into an unexplored area where no absolute or relative dating techniques are available. There he finds a number of sites with potsherds of types new to archaeology. Can he date these sites? Again he uses seriation, but this time he inverts the logical proposition. If things which look alike tend to be of the same age, things which do not look alike should tend to be of different ages. His first task would be to separate all the different types of potsherds. Taking a specimen of each type, he lays them out in a row. If there is any difference in the sites through time, the styles of pottery will change correspondingly. But there will be some styles which change slowly and some which will influence others. For example, if we were to seriate the style of automobile rear fenders for the period 1956 to 1964, we would observe that they began to sprout taller and taller fins; then the size of the fins became reduced more and more. Some of the style aspects of pre-1956 rear fenders went along with the development of fins and others dropped out. Some of the style aspects which were developed with the fins were retained after the fins decreased in size.