diff --git a/paper/paper.md b/paper/paper.md
index 20716b9..c0d1a2e 100644
--- a/paper/paper.md
+++ b/paper/paper.md
@@ -169,16 +169,19 @@ UserIndividual = MonoIndividual[BinaryChromosome // n].set_fitness(lambda o: _ev
 
 or with the helper:
 UserIndividual = makeIndividual(n_chromosomes=1, size=n).set_fitness(lambda o: _evaluate(o.chromosome))
-
-or let the individual be a chromosome:
-UserIndividual = (BinaryChromosome // n).set_fitness(lambda o: _evaluate(o.chromosome))
 """
 
 UserPopulation = StandardPopulation[UserIndividual] // 20
-"""
 ```
 
-We can collect all the codes to one line:
+You also see that the equivalent expressions no longer explicitly depends on class inheritance, making the code more concise and similar to algebraic operation.
+
+We can also consider chromosomes as the elements of the population, and define an individual as follows:
+```
+UserIndividual = (BinaryChromosome // n).set_fitness(lambda o: _evaluate(o.chromosome))
+```
+
+To further streamline the code, we integrate all the components into a single line:
 ```UserPopulation = StandardPopulation[BinaryChromosome // n].set_fitness(_evaluate)```
 
 
@@ -189,9 +192,6 @@ pop = UserPopulation.random()
 pop.evolve(n_iter=100)
 ```
 
-You also see that the equivalent expressions no longer explicitly depends on class inheritance, making the code more concise and similar to algebraic operation.
-
-
 # Visualization
 
 Instead of implementing visualization methods, `pyrimidine` yields a `pandas.DataFrame` object that encapsulates statistical results for each generation by setting `history=True` in `evolve` method. Users can harness this object to plot the performance curves. Generally, users are required to furnish a "statistic dictionary" whose keys are the names of the statistics, and values are functions mapping the population to numerical values (strings are confined to pre-defined methods or attributes of the population).