A charge half the width of a bend is a bendlet (Fig. 80), and one half the width of a bendlet is a cottise (Fig. 81), but a cottise cannot exist alone, inasmuch as it has of itself neither direction nor position, but is only found accompanying one of the ordinaries. The arms of Harley are an example of a bend cottised. Bendlets will very seldom be found either in addition to a bend, or charged, but the arms of Vaile show both these peculiarities. [Illustration: FIG. 80.--Bendlets.] A bend will usually be found between two charges. Occasionally it will be found between four, but more frequently between six. In none of these cases is it necessary to specify the position of the subsidiary charges. It is presumed that the bend separates them into even numbers, but their exact position (beyond this) upon the shield is left to the judgment of the artist, and their disposition is governed by the space left available by the shape of the shield. A further presumption is permitted in the case of a bend between _three_ objects, which are presumed to be two in chief and one in base. But even in the case of three the position will be usually found to be specifically stated, as would be the case with any other uneven number. [Illustration: FIG. 81.--Bend cottised.] Charges on a bend are placed in the direction of the bend. In such cases it is not necessary to specify that the charges are bendwise. When a charge or charges occupy the position which a bend would, they are said to be placed "in bend." This is not the same thing as a charge placed "bendwise" (or bendways). In this case the charge itself is slanted into the angle at which the bend crosses the shield, but the position of the charge upon the shield is not governed thereby. When a bend and chief occur together in the same arms, the chief will usually surmount the bend, the latter issuing from the angle between the base of the chief and the side of the shield. An instance to the contrary, however, will be found in the arms of Fitz-Herbert of Swynnerton, in which the bend is continued over the chief. This instance, however (as doubtless all others of the kind), is due to the {114} use of the bend in early times as a mark of difference. The coat of arms, therefore, had an earlier and separate existence without the bend, which has been superimposed as a difference upon a previously existing coat. The use of the bend as a difference will be again referred to when considering more fully the marks and methods of indicating cadency. [Illustration: FIG. 82.--Bend sinister.] A curious instance of the use of the sun's rays in bend will be found in the arms of Warde-Aldam.[8]