172 
OF THE FERNS AND FLOWERING PLANTS. 
plant is absorbed by the sand. It is then removed with | ■ 
great care, and flattened in the usual manner, between 
sheets of paper. By this plan the petals retain their colour fl 
in a way that I have never seen equalled. — Ed.] j 
With regard to the physiological and anatomical inves- 
tigation of the flowering plants, the limits of this little ai 
volume will not allow of my entering into it, nor would it ai 
lie within my province to do so. 
[At the same time I offer no apology for concluding this 
chapter with the following useful suggestions from the late 
Dr. Lindley’s excellent 1 Descriptive Botany.’ 
1 The student should select for examination as perfect a 
specimen as he can obtain, and should carefully study every 
part. ... In doing this he must on no account guess, 
but be certain that he sees correctly, what is before him. 
This is not difficult in the case of roots, stems, leaves, and 
their parts ; but the flower, from its general smallness and 
somewhat complicated structure, demands a little skill in 
dissection, which is only to be gained by experience. 
‘After its external structure has been determined, it is 
necessary to open the flower. Mere looking down into its 
tube or interior leads to nothing but error. The student 
should hold it in his left hand, and split it longitudinally by 
a rapid cut from below upwards. This lays bare the whole 
of the interior, shows the number and position of their 
parts, and their insertion, which is very important. If he 
attempts to divide a flower by cutting it from above down- 
wards, he only crushes and disfigures his specimen. In the 
case of the ovary it is usually necessary to ascertain its 
placentation, which, if it is not seen in the first longitudinal 
section, can be best determined by making a transverse 
section. ... In examining seeds of any kind, where dis- 
section is required, cut into them perpendicularly, beginning 
at the hilum, and passing the knife through the axis : m 
this way the embryo and its relation to other parts usually 
^ becomes distinctly visible. Failing this, the observer mns 
